[{"content":" SyCoSMA #SyCoSMA is a research team at the LIRIS laboratory in Lyon, France.\nThe research conducted by the SyCoSMA team is structured around two main axes: firstly, multi-agent learning and coordination; secondly, the cognitive development of an agent through the construction of an internal representation of its environment.\n","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/","section":"","summary":"","title":""},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-05694975/","section":"Publications","summary":"","title":"2026 GTAS (Working Group on Animation and Simulation) workshop in Lyon. Tutorial on human motion capture tools from monocular RGB videos and their Neural Network-Based Classification through action categories"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-05603249/","section":"Publications","summary":"","title":"A Survey of Multi-Agent Deep Reinforcement Learning with Graph Neural Network-Based Communication"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/alexandre-meyer/","section":"Authors","summary":"","title":"Alexandre Meyer"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/tags/argumentation/","section":"Tags","summary":"","title":"Argumentation"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/","section":"Authors","summary":"","title":"Authors"},{"content":"The SyCoSMA team is organized along 2 axes.\nArtificial Cognitive Systems This axis addresses questions of intelligence, the intelligibility of knowledge, the explicability of processes and the ethics of decisions. Inspired by Piaget\u0026rsquo;s work on developmental learning, the SyCoSMA team is exploring how an agent, with no a priori knowledge, can learn a predictive sensorimotor model of the world. Developed incrementally and autonomously, this model enables the agent to anticipate its actions and adapt to changes in its environment.\nMulti-Agent Systems Multi-agent coordination models target collective problem-solving. To this end, the SyCoSMA team is developing techniques for argumentation, automatic negotiation and the generation of coalition structures to enable scaling up. The team also combines the techniques of multi-agent deep reinforcement learning and developmental learning to make the exploration process more robust.\nThe various projects the team leads, or is involved in, listed below, belong to one or both of these axes.\n","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/projects/","section":"Axes","summary":"","title":"Axes"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/bruno-yun/","section":"Authors","summary":"","title":"Bruno Yun"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-05702375/","section":"Publications","summary":"","title":"CLIMB: Centroid-Based Hierarchical Memory for Online Continual Self-Supervised Learning"},{"content":" Scientific lead: CHAPUT Rémy Is SyCoSMA the project lead? Yes Start date: 2026-01-01 End date: 2027-12-31 Scientific partner: LIMOS With the increasing use of Artificial Intelligence (AI), it is more important than ever to ensure that these systems behave in a way that is aligned with our moral values, to guarantee ben- eficial use. This requires several capabilities from the AI system: being able to represent moral values; being able to exhibit a behavior that takes these values into account; and being able to ex- plain the behavior to humans (regulators, users, or even stakeholders) to verify that the moral values are indeed respected.\nThe ECIÉA project (Explication du Comportement d\u0026rsquo;un agent IA Éthiquement Aligné \u0026ndash; Explaining the behavior of an ethically-aligned agent) proposes to implement these capabilities through a hybrid combination of abstract argumentation and Reinforce- ment Learning (RL). RL will provide adaptability to agents, but requires a reward function to guide the learning towards the desired behavior. Argumentation is an interesting framework for representing these moral values, as it is fairly close to human reasoning.\nUsing argumentation to design the reward functions of learning agents will allow to include non-AI experts in the design process, and explain the resulting behavior in terms of rewards (i.e., what drives the agent towards this specific behavior).\n","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/projects/32_eciea/","section":"Axes","summary":"","title":"ECIÉA"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/emmanuel-dellandr%C3%A9a/","section":"Authors","summary":"","title":"Emmanuel Dellandréa"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/tags/ethics/","section":"Tags","summary":"","title":"Ethics"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-05601221/","section":"Publications","summary":"","title":"Evaluating Social Credibility of Generative Agents: Incentives, Beliefs, and Coordination"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-05676864/","section":"Publications","summary":"","title":"Évaluation de la crédibilité sociale des agents génératifs : incitations, croyances et coordination"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-05669745/","section":"Publications","summary":"","title":"Exploration and Online Transfer with Behavioral Foundation Models"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/farah-ben-slama/","section":"Authors","summary":"","title":"Farah Ben Slama"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/fr%C3%A9d%C3%A9ric-armetta/","section":"Authors","summary":"","title":"Frédéric Armetta"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/gowrishankar-ganesh/","section":"Authors","summary":"","title":"Gowrishankar Ganesh"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-05618953/","section":"Publications","summary":"","title":"Instance Discrimination for Link Prediction"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/julien-lefebvre/","section":"Authors","summary":"","title":"Julien Lefebvre"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/laetitia-matignon/","section":"Authors","summary":"","title":"Laetitia Matignon"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/la%C3%ABtitia-matignon/","section":"Authors","summary":"","title":"Laëtitia Matignon"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-05447222/","section":"Publications","summary":"","title":"Large Language Models and Algorithm Execution: Application to an Arithmetic Function"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/louis-bagot/","section":"Authors","summary":"","title":"Louis Bagot"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/madalina-croitoru/","section":"Authors","summary":"","title":"Madalina Croitoru"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/mathieu-guillermin/","section":"Authors","summary":"","title":"Mathieu Guillermin"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/mathieu-lefort/","section":"Authors","summary":"","title":"Mathieu Lefort"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/maxime-morge/","section":"Authors","summary":"","title":"Maxime Morge"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/michael-a-m%C3%BCller/","section":"Authors","summary":"","title":"Michael a Müller"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/nathan-salazar/","section":"Authors","summary":"","title":"Nathan Salazar"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/nicolas-pronost/","section":"Authors","summary":"","title":"Nicolas Pronost"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/","section":"Publications","summary":"","title":"Publications"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-05568781/","section":"Publications","summary":"","title":"Qu’est-ce que l’« IA agentique » ? Comprendre son histoire pour dépasser l’effet de mode"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/tags/reinforcement-learning/","section":"Tags","summary":"","title":"Reinforcement Learning"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-05540290/","section":"Publications","summary":"","title":"Rejecting Arguments Based on Doubt in Structured Bipolar Argumentation"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/r%C3%A9my-cazabet/","section":"Authors","summary":"","title":"Rémy Cazabet"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/r%C3%A9my-chaput/","section":"Authors","summary":"","title":"Rémy Chaput"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/sandra-victor/","section":"Authors","summary":"","title":"Sandra Victor"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-05406402/","section":"Publications","summary":"","title":"Socially aware robot for reward, effort, and risk distribution"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/srdjan-vesic/","section":"Authors","summary":"","title":"Srdjan Vesic"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/stefan-duffner/","section":"Authors","summary":"","title":"Stefan Duffner"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/st%C3%A9phane-bonnevay/","section":"Authors","summary":"","title":"Stéphane Bonnevay"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/tags/","section":"Tags","summary":"","title":"Tags"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/timon-deschamps/","section":"Authors","summary":"","title":"Timon Deschamps"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-05627554/","section":"Publications","summary":"","title":"Utility-Based Soft Masking for Continual Multi-Objective Reinforcement Learning"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/valentin-cuzin-rambaud/","section":"Authors","summary":"","title":"Valentin Cuzin-Rambaud"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-05666039/","section":"Publications","summary":"","title":"XmoPipe: A Pipeline for Large-Scale In-the-Wild Human Motion Dataset Construction"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-04828393/","section":"Publications","summary":"","title":"A Multiagent Path Search Algorithm for Large-Scale Coalition Structure Generation, Accepted Paper"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-05391808/","section":"Publications","summary":"","title":"A Reinforcement Learning Simulator for Multi-UAV Based Network Coverage Problem"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-05324845/","section":"Publications","summary":"","title":"Action matters for object representation learning from sequences"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/alexandre-poisson/","section":"Authors","summary":"","title":"Alexandre Poisson"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/alexis-le-pichon/","section":"Authors","summary":"","title":"Alexis Le Pichon"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/alice-janela-cameijo/","section":"Authors","summary":"","title":"Alice Janela Cameijo"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-05769011/","section":"Publications","summary":"","title":"AMELIA: A Family of Multi-task End-to-end Language Models for Argumentation"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/amit-biswas/","section":"Authors","summary":"","title":"Amit Biswas"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/anil-kumar-tripathi/","section":"Authors","summary":"","title":"Anil Kumar Tripathi"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/animesh-dutta/","section":"Authors","summary":"","title":"Animesh Dutta"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/antoine-richard/","section":"Authors","summary":"","title":"Antoine Richard"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/axel-bessy/","section":"Authors","summary":"","title":"Axel Bessy"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/biswas-tuhin-kumar/","section":"Authors","summary":"","title":"Biswas Tuhin Kumar"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/brooke-stephenson/","section":"Authors","summary":"","title":"Brooke Stephenson"},{"content":" Scientific lead: YUN Bruno Is SyCoSMA the project lead? Yes Start date: 2025-01-01 End date: 2025-12-31 TThis project has two objectives:\n(OB1) Enable the automatic mapping of debates into structured format via lightweight LLMs. This objective will require the creation of a multi-task argumentation dataset, the training of a custom LLM model, and its usage on argumentation graphs to assist users. (OB2) Provide hybrid argumentation semantics to exploit the data of structured debates. The idea is to explore how argument quality dimensions can be used to create new semantics and how they perform in real-life. ","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/projects/31_cade-ai/","section":"Axes","summary":"","title":"CADE-AI - Collaborative Argumentation and Debate Enhancement with AI"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/caren-al-anaissy/","section":"Authors","summary":"","title":"Caren Al Anaissy"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-04969435/","section":"Publications","summary":"","title":"Compact agent neighborhood search for the SCSGA-MF-TS: SCSGA with multi-dimensional features prioritizing task satisfaction"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/constantino-listowski/","section":"Authors","summary":"","title":"Constantino Listowski"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/tags/cooperation-and-coordination/","section":"Tags","summary":"","title":"Cooperation and Coordination"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/tags/debates/","section":"Tags","summary":"","title":"Debates"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/djamila-boukredera/","section":"Authors","summary":"","title":"Djamila Boukredera"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/dorian-tonnis/","section":"Authors","summary":"","title":"Dorian Tonnis"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/emilien-komlenovic/","section":"Authors","summary":"","title":"Emilien Komlenovic"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/tags/evaluation-of-generative-agents/","section":"Tags","summary":"","title":"Evaluation of Generative Agents"},{"content":" Scientific lead: MORGE Maxime Is SyCoSMA the project lead? Yes Start date: 2025-01-01 End date: 2027-12-31 Scientific partner: ERIC (LYON2) Abstract #This project explores the potential of Generative Autonomous Agents and Multiagent Systems (GAAMAS) for social simulation. It aims to better understand how these artificial entities, powered by Large Language Models (LLMs), interact, make decisions, adapt to others\u0026rsquo; behaviour, and simulate human reasoning, particularly in strategic contexts inspired by Game Theory. This project will contribute to assessing the current capabilities and limitations of GAAMAS and to proposing concrete avenues for improving their coherence and realism in social simulations.\nRésumé #Ce projet explore le potentiel des agents autonomes et des systèmes multi-agents génératifs (Generative Autonomous Agents and Multiagent Systems, GAAMAS) pour la simulation sociale. Il vise à mieux comprendre comment ces entités artificielles, propulsées par les grands modèles de langage (Large Language Models, LLM), interagissent, prennent des décisions, s’adaptent aux comportements d’autrui et simulent des raisonnements humains, notamment dans des contextes stratégiques inspirés de la théorie des jeux. Ce projet contribuera à évaluer les capacités et les limites actuelles des GAAMAS, et à proposer des pistes concrètes pour améliorer leur cohérence et leur réalisme dans les simulations sociales.\nPublications #Maxime Morge. Système multi-agents et agents autonomes génératifs : état de l\u0026rsquo;art des connaissances et des ressources. Trente-troisièmes journées francophones sur les systèmes multi-agents (JFSMA), Jul 2025, Dijon, France. pp.79-88. hal-05146426\nMaxime Morge. Is Generative Artificial Intelligence Ready for Computational Social Science? ⋆. Advances in Practical Applications of Agents, Multi-Agent Systems, and Complex Systems Simulation. The PAAMS Collection, Mathieu, Philippe and Corchado, Juan Manuel and De La Prieta, Fernando, Jun 2025, Lille, France. hal-05132301\nAcknowledgments #The GAAMAS project (Generative Autonomous Agents and Multi-Agent Systems) received financial support from the Rhône-Alpes Institute of Complex Systems and from the Lyon Computer Science Federation.\n","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/projects/35_gaamas/","section":"Axes","summary":"","title":"GAAMAS"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/gaurav-baranwal/","section":"Authors","summary":"","title":"Gaurav Baranwal"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/g%C3%A9rard-bailly/","section":"Authors","summary":"","title":"Gérard Bailly"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-05245062/","section":"Publications","summary":"","title":"Grands modèles de langage et exécution d'algorithmes. Application à une fonction arithmétique."},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/tags/graph/","section":"Tags","summary":"","title":"Graph"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/gupta-avisek/","section":"Authors","summary":"","title":"Gupta Avisek"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/hamid-ladjal/","section":"Authors","summary":"","title":"Hamid Ladjal"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/henri-savigny/","section":"Authors","summary":"","title":"Henri Savigny"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-04828421/","section":"Publications","summary":"","title":"Hide Exposures by Removing Mastermind’s External Sources on Social Network (Student Abstract)"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/hugo-imbert/","section":"Authors","summary":"","title":"Hugo Imbert"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-05525396/","section":"Publications","summary":"","title":"Human compliance with computational argumentation principles"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-05348749/","section":"Publications","summary":"","title":"Identifying and Localizing Dynamic Affordances to Improve Interactions with Other Agents in Continuous Environments"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-04857142/","section":"Publications","summary":"","title":"Impact Measures for Gradual Argumentation Semantics"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-05590807/","section":"Publications","summary":"","title":"Interpreting Preferred Semantics in Structured Bipolar Argumentation"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-05132301/","section":"Publications","summary":"","title":"Is Generative Artificial Intelligence Ready for Computational Social Science? ⋆"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/isabelle-gu%C3%A9rin-lassous/","section":"Authors","summary":"","title":"Isabelle Guérin Lassous"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/jean-charles-quinton/","section":"Authors","summary":"","title":"Jean-Charles Quinton"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/jean-paul-jamont/","section":"Authors","summary":"","title":"Jean-Paul Jamont"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/j%C3%A9r%C3%B4me-delobelle/","section":"Authors","summary":"","title":"Jérôme Delobelle"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/tags/large-language-models/","section":"Tags","summary":"","title":"Large Language Models"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-05245220/","section":"Publications","summary":"","title":"Leveraging Graph Structures and Large Language Models for End-to-End Synthetic Task-Oriented Dialogues"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-05408360/","section":"Publications","summary":"","title":"Leveraging Latent Space and Radiomics for Multi-Label Classification of Chest CT Scans"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/tags/link-prediction/","section":"Tags","summary":"","title":"Link Prediction"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/lo%C3%AFc-desgeorges/","section":"Authors","summary":"","title":"Loïc Desgeorges"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/lucas-fournier/","section":"Authors","summary":"","title":"Lucas Fournier"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/manisha-singh/","section":"Authors","summary":"","title":"Manisha Singh"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/maya-medjad/","section":"Authors","summary":"","title":"Maya Medjad"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/micha%C3%ABl-sdika/","section":"Authors","summary":"","title":"Michaël Sdika"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/mohamed-massamba-sene/","section":"Authors","summary":"","title":"Mohamed Massamba Sene"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-04943542/","section":"Publications","summary":"","title":"Multi-attribute based self-stabilizing algorithm for leader election in distributed systems"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/narayan-changder/","section":"Authors","summary":"","title":"Narayan Changder"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/nicolas-duchateau/","section":"Authors","summary":"","title":"Nicolas Duchateau"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/nilanjana-saha/","section":"Authors","summary":"","title":"Nilanjana Saha"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/olivier-georgeon/","section":"Authors","summary":"","title":"Olivier Georgeon"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/olivier-perrotin/","section":"Authors","summary":"","title":"Olivier Perrotin"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-05324750/","section":"Publications","summary":"","title":"Online Sensorimotor Sequence-based Learning Using Predictive Trees"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/pierre-elliott-thiboud/","section":"Authors","summary":"","title":"Pierre-Elliott Thiboud"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/predrag-teovanovi%C4%87/","section":"Authors","summary":"","title":"Predrag Teovanović"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/quentin-brissaud/","section":"Authors","summary":"","title":"Quentin Brissaud"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/rapha%C3%ABl-szymocha/","section":"Authors","summary":"","title":"Raphaël Szymocha"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/redha-taguelmimt/","section":"Authors","summary":"","title":"Redha Taguelmimt"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-04813569/","section":"Publications","summary":"","title":"Refining the evaluation of speech synthesis: A summary of the Blizzard Challenge 2023"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-04822315/","section":"Publications","summary":"","title":"Review of driver behaviour modelling for highway on‐ramp merging"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/samir-aknine/","section":"Authors","summary":"","title":"Samir Aknine"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/tags/self-supervised-learning/","section":"Tags","summary":"","title":"Self-Supervised Learning"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/silvain-gerber/","section":"Authors","summary":"","title":"Silvain Gerber"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/simon-achard/","section":"Authors","summary":"","title":"Simon Achard"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/simon-gay/","section":"Authors","summary":"","title":"Simon Gay"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/simon-king/","section":"Authors","summary":"","title":"Simon King"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/souhila-arib/","section":"Authors","summary":"","title":"Souhila Arib"},{"content":" Scientific lead: CAZABET Rémy Is SyCoSMA the project lead? No Start date: 2025-01-01 End date: 2025-12-01 Le projet visera à explorer l’utilisation de l’apprentissage auto-supervisé pour un problème classique en science des réseaux, la prédiction de lien. L’objectif de cette tâche est de prédire, pour un graphe donné, quelles sont les paires de nœuds ayant le plus de chance d’être connectées par un lien, parmi celles qui ne le sont pas déjà. Au-delà d’applications pratiques telles que la recommandation (e.g., lien utilisateur/produit) ou la santé (interactions entre médicaments ou entre protéines), la capacité à prédire un lien est également un proxy pour évaluer notre capacité à comprendre comment un graphe est organisé : l’organisation du graphe est en effet intégralement définie par ses liens, et réciproquement, les liens observés sont une conséquence de l’organisation du graphe. Dans les approches classiques de prédiction de lien par GNN, la prédiction d’un lien entre deux nœuds se calcule à partir des paires de vecteurs des nœuds correspondant dans l’espace de plongement de ce graphe. L’objectif des méthodes d’apprentissage contrastif dans ce contexte est d’être capable de guider le plongement des nœuds de manière à préserver des propriétés désirées, tout en en variant d’autres. Par exemple, un modèle de graphe en bloc (Stochastic Block Model, SBM) ayant une structure en communauté donnée peut générer un grand nombre de graphes différents, ayant peu de liens en commun, mais correspondant tous à une même organisation. Cette information pourrait être utilisée, en générant des paires de graphes ayant peu de liens en commun, mais une même structure, ou au contraire de nombreux liens en commun, mais des structures différentes. Des travaux dans cette direction ont été publiés récemment. Nous commencerons par reproduire ces résultats, afin de les évaluer dans le contexte qui nous intéresse, celui de la prédiction de lien. Nous poursuivrons en proposant des approches originales. En particulier, la plupart des approches existantes ne cherchent qu’à conserver les structures en communauté du graphe. Nous chercherons à préserver un ensemble de structures, en générant des paires contrastives non seulement pour des graphes différents ayant la même structure en communauté, mais également des paires de graphes ayant la même structure spatiale, en cœur périphérie, etc. Nous pourrons nous appuyer sur des travaux récents de l’équipe SyCoSMA utilisant des techniques similaires sur des images. En imposant un plongement préservant autant d’aspects que possible de la structure du réseau, nous pensons être en mesure d’améliorer la capacité à capturer la complexité de son organisation, et donc à prédire ses futurs liens.\n","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/projects/34_ssl4lp/","section":"Axes","summary":"","title":"SSL4LP"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/sven-peter-n%C3%A4sholm/","section":"Authors","summary":"","title":"Sven Peter Näsholm"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/swagatam-das/","section":"Authors","summary":"","title":"Swagatam Das"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-05146426/","section":"Publications","summary":"","title":"Système multi-agents et agents autonomes génératifs : état de l'art des connaissances et des ressources"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-05291344/","section":"Publications","summary":"","title":"SysTemp : Système multi-agents par la génération basée sur un templatede squelettes pour le SysML v2"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-05249466/","section":"Publications","summary":"","title":"SysTemp: A Multi-Agent System for Template-Based Generation of SysML v2"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/thomas-barba/","section":"Authors","summary":"","title":"Thomas Barba"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-05263060/","section":"Publications","summary":"","title":"Towards real-time assessment of infrasound event detection capability using deep learning-based transmission loss estimation"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/tuomas-sandholm/","section":"Authors","summary":"","title":"Tuomas Sandholm"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/valentine-wargnier-dauchelle/","section":"Authors","summary":"","title":"Valentine Wargnier-Dauchelle"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-04971677/","section":"Publications","summary":"","title":"Vers une meilleure interprétabilité des réseaux de neurones pour la prédiction du sepsis via des cartes d'attribution variable par variable"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-05368143/","section":"Publications","summary":"","title":"VIRAL: VISION-GROUNDED INTEGRATION FOR REWARD DESIGN AND LEARNING"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/yasmine-bouamra/","section":"Authors","summary":"","title":"Yasmine Bouamra"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/youcef-sklab/","section":"Authors","summary":"","title":"Youcef Sklab"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/zine-el-abidine-kherroubi/","section":"Authors","summary":"","title":"Zine El Abidine Kherroubi"},{"content":" Scientific lead: MEYER Alexandre Is SyCoSMA the project lead? No Start date: 2024-01-12 End date: 2027-01-12 L\u0026rsquo;imagerie thoracique est souvent limitée à des diagnostics spécifiques. Ce projet cherche à améliorer les algorithmes de détection afin de traiter plusieurs pathologies simultanément.\n","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/projects/8_atos/","section":"Axes","summary":"","title":"ATOS"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/tags/continual-learning/","section":"Tags","summary":"","title":"Continual Learning"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/tags/medical-scanner-data/","section":"Tags","summary":"","title":"Medical Scanner Data"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/tags/multi-label-classification/","section":"Tags","summary":"","title":"Multi-Label Classification"},{"content":" Scientific lead: LEFORT Mathieu Is SyCoSMA the project lead? Yes Start date: 2024-01-12 End date: 2027-01-12 L’apprentissage auto-supervisé s’est imposé comme une approche clé en intelligence artificielle, permettant aux modèles d’apprendre des représentations pertinentes à partir de données non annotées. Cette approche s’avère particulièrement prometteuse dans le domaine de la vision par ordinateur, où l’annotation manuelle représente un coût considérable. Nos travaux s’inscrivent dans le domaine des méthodes d’apprentissage auto-supervisé non génératives. Dans le cadre d’un agent autonome, l’apprentissage continu (incrémentale ou lifelong) est une nécessité fondamentale. Un tel agent, évoluant dans un environnement dynamique et interagissant physiquement avec celui-ci, doit constamment adapter ses représentations internes face aux nouvelles tâches, tout en préservant les connaissances précédemment acquises. L’apprentissage auto-supervisé (SSL) offre un cadre particulièrement prometteur pour ces agents, en éliminant le besoin d’annotations externes. Le domaine de l’apprentissage continu auto-supervisé est relativement actif, avec des travaux récents tels que CaSSLe [1], Osiris [2] et Branch-Tuning [3]. Le point commun entre ces travaux est l’utilisation de connaissances externes pour guider l’apprentissage, en modifiant le modèle à chaque changement de tâche. Cependant, les approches continues, où les données ne sont montrées qu’une seule fois au modèle et traitées de manière séquentielle, où les transitions entre les tâches ne sont pas connues, sont encore peu explorées en SSL. Dans ce domaine, les tâches sont généralement générées en découpant des jeux de données tels qu’ImageNet en groupes de classes, sans intersection entre deux tâches. Dans l’apprentissage incrémental il existe un sous domaine qui est l’apprentissage online où les données sont présentées une seule fois au modèle. Notre travail s’inscrit dans cette problématique : l’apprentissage continu auto-supervisé en ligne, sans connaissance préalable des transitions.\n","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/projects/10_sens_lefort/","section":"Axes","summary":"","title":"SENS LEFORT"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/tags/active-learning/","section":"Tags","summary":"","title":"Active Learning"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/tags/active-perception/","section":"Tags","summary":"","title":"Active Perception"},{"content":" Scientific lead: LEFORT Mathieu Is SyCoSMA the project lead? Yes Start date: 2023-01-10 End date: 2028-03-31 In deep learning, many recent models of Self-Supervised Learning (SSL) use instance based discriminative tasks. The underlying idea is to define positive (resp. negative) pairs of data whose representations have to be similar (resp. dissimilar). To define these positive pairs, the models consider a set of augmentations (e.g. resize, color jitter, blur, etc.) to be applied to the same inputs. The models learn to be invariant (or equivariant where considering the impact of the augmentation on the representation) to the properties modified by these augmentations, but not to the semantic content of the input.\nHowever, there is “a fundamental misalignment between human and typical AI representations: while the former are grounded in rich sensorimotor experience, the latter are typically passive and limited to a few modalities such as vision and text”. In this project, we propose to take inspiration from the way babies learn to explore their environment through actions that shape their multimodal experience to improve. We will build more specifically upon sensorimotor contingencies theory, which combines coherent pieces of evidence from neuroscience and psychology in a unified framework. The key claims are about:\nregularities, with sensory motor contingencies defined as “the structure of the rules governing the sensory changes produced by various motor actions” (e.g. a straight line will be represented as the perceptual invariance caused by the eyes translation along this line) active perception as the “organism’s exploration of the environment that is mediated by knowledge of sensory motor contingencies”. Our claim in this project is that to go beyond invariant and equivariant pretext tasks, the next research axis to explore in SSL is to consider action as the core of the representation learning and perception process, guided by the concepts of sensorimotor contingencies theory. Thus, this conceptual shift of using action as an unifying key point in learning will guide towards more general architectures and representations. It will also allow to interact with the environment to have access to all its dynamic and properties. Moreover, having more human like perceptual and learning mechanisms should help to generalize to various environments, as human do.\nProject website\n","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/projects/21_me_smerise/","section":"Axes","summary":"","title":"MeSMRise"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/tags/multimodal-fusion/","section":"Tags","summary":"","title":"Multimodal Fusion"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/tags/sensorimotor-theories/","section":"Tags","summary":"","title":"Sensorimotor Theories"},{"content":" Scientific lead: LEFORT Mathieu Is SyCoSMA the project lead? Yes Start date: 2024-03-15 End date: 2027-12-09 Dans un monde où l\u0026rsquo;efficacité technologique et l\u0026rsquo;innovation sont primordiales, Neovision et le LIRIS unissent leurs forces pour transformer le développement de l\u0026rsquo;intelligence artificielle (IA) dans les entreprises. Notre mission est de rendre ce développement plus rapide, plus précis et plus durable d\u0026rsquo;un point de vue technique et environnemental. En nous appuyant sur les dernières avancées en matière d\u0026rsquo;apprentissage auto-supervisé, nous améliorons Tadaviz, un outil d\u0026rsquo;annotation avancé développé par Neovision, en augmentant la qualité et la flexibilité du processus d\u0026rsquo;annotation des données. Notre approche frugale de l\u0026rsquo;IA vise à réduire les ressources nécessaires tout en minimisant l\u0026rsquo;impact sur l\u0026rsquo;environnement. Avec l\u0026rsquo;intégration de fonctions de recherche textuelle, Tadaviz devient encore plus puissant, permettant une navigation intuitive dans des ensembles de données complexes. Les entreprises utilisant des technologies similaires ont constaté une réduction de 50 % du temps de développement de l\u0026rsquo;IA, tout en améliorant la précision des modèles de plus de 5%. En mettant le cap sur ce projet, nous faisons le choix de l\u0026rsquo;innovation durable, de l\u0026rsquo;efficacité et de la technologie de pointe, pour façonner ensemble l\u0026rsquo;avenir de l\u0026rsquo;intelligence artificielle.\n","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/projects/9_datawise/","section":"Axes","summary":"","title":"DATAWISE"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/tags/frugal-ai/","section":"Tags","summary":"","title":"Frugal AI"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/tags/transfer-learning/","section":"Tags","summary":"","title":"Transfer Learning"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/tags/agent-behaviour/","section":"Tags","summary":"","title":"Agent Behaviour"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/tags/agent-based-simulation/","section":"Tags","summary":"","title":"Agent-Based Simulation"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/tags/ai-assistance/","section":"Tags","summary":"","title":"AI Assistance"},{"content":" Scientific lead: ARMETTA Is SyCoSMA the project lead? Yes Start date: 2024-01-10 End date: 2027-01-01 As part of this PhD work, we aim to develop an assistant to support the design of complex technological tools—a field in which Siemens offers a wide range of software products. This design process typically relies on the SysML v2 language and various dedicated tools (Model-Based Systems Engineering, or MBSE), and requires extensive training and significant experience to be fully mastered.\nThe objective of this thesis is therefore to develop intelligent assistance for designers, based on specifications and/or descriptive data expressed in natural language.\n","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/projects/12_siemens_industry_software/","section":"Axes","summary":"","title":"Assistance in designing complex technological tools"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/tags/complex-technological-tools/","section":"Tags","summary":"","title":"Complex Technological Tools"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/tags/llm/","section":"Tags","summary":"","title":"LLM"},{"content":" Scientific lead: Maxime MORGE Is SyCoSMA the project lead? No Start date: 2024-01-10 End date: 2027-12-31 Scientific partner: CLERSÉ (ULille), CRIStAL (ULille) Adopting an interdisciplinary approach combining sociology and computer science, this project seeks to understand the social dynamics that result from individual behaviour in online communities. It focus on communities of data analysts who share an interest in statistical analysis tools (e.g. R or Pandas) on the Stack Overflow platform. The programme emphasises exploration and an inductive approach in order to lay the foundations for theorising and agent-based modelling.\nThe RORES-CL project receives financial support from the CNRS through the MITI interdisciplinary programmes. It is supported by the complementary expertise of CLERSÉ (Centre Lillois d\u0026rsquo;Études et de Recherches Sociologiques et Économiques) at University of Lille and the LIRIS laboratory at Claude Bernard Lyon 1 University.\n","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/projects/6_rores-cl/","section":"Axes","summary":"","title":"RORES-CL - Rôles et réseaux sociaux dans les communautés en ligne."},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/tags/social-network-analysis/","section":"Tags","summary":"","title":"Social Network Analysis"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-04357674/","section":"Publications","summary":"","title":"A Multiagent Path Search Algorithm for Large-Scale Coalition Structure Generation, Long Abstract"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-04903654/","section":"Publications","summary":"","title":"Adaptive Learning en contexte parascolaire : comprendre les usages et effets via l'analyse des traces d'un déploiement industriel"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-04662012/","section":"Publications","summary":"","title":"ADBL2: Assisted Debate Builder with Large Language Models"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/ana%C3%ABlle-badier/","section":"Authors","summary":"","title":"Anaëlle Badier"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/angela-bonifati/","section":"Authors","summary":"","title":"Angela Bonifati"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/anne-c%C3%A9cile-caron/","section":"Authors","summary":"","title":"Anne-Cécile Caron"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-05349584/","section":"Publications","summary":"","title":"Argumentation-based applications for decision-making"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-04587508/","section":"Publications","summary":"","title":"Artificial Enactive Inference in Three-Dimensional World"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-04742709/","section":"Publications","summary":"","title":"Assisted Debated Builder with Large Language Models"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-04427749/","section":"Publications","summary":"","title":"Attention Graph for Multi-Robot Social Navigation with Deep Reinforcement Learning"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/tags/augmented-reality/","section":"Tags","summary":"","title":"Augmented Reality"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-04688608/","section":"Publications","summary":"","title":"AutoNeRF: Training Implicit Scene Representations with Autonomous Agents"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/tel-04842891/","section":"Publications","summary":"","title":"Autonomie et représentation de l'environnement"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/b%C3%A9atrice-de-mont%C3%A9ra/","section":"Authors","summary":"","title":"Béatrice De Montéra"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/chattopadhyay-samiran/","section":"Authors","summary":"","title":"Chattopadhyay Samiran"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/christian-wolf/","section":"Authors","summary":"","title":"Christian Wolf"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/tags/co-creation/","section":"Tags","summary":"","title":"Co-Creation"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-04418100/","section":"Publications","summary":"","title":"Coalition Formation for Task Allocation Using Multiple Distance Metrics (Student Abstract)"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/danka-puri%C4%87/","section":"Authors","summary":"","title":"Danka Purić"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/david-lurie/","section":"Authors","summary":"","title":"David Lurie"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/tags/decision-making/","section":"Tags","summary":"","title":"Decision Making"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-04842264/","section":"Publications","summary":"","title":"Deep learning methods for modeling infrasound transmission loss in the middle atmosphere"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/devendra-singh-chaplot/","section":"Authors","summary":"","title":"Devendra Singh Chaplot"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/dhruv-batra/","section":"Authors","summary":"","title":"Dhruv Batra"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-04357671/","section":"Publications","summary":"","title":"Efficient Size-based Hybrid Algorithm for Optimal Coalition Structure Generation, Long Abstract"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/ellie-beauprez/","section":"Authors","summary":"","title":"Ellie Beauprez"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/elliot-ambroise-faugier/","section":"Authors","summary":"","title":"Elliot Ambroise Faugier"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/elliot-faugier/","section":"Authors","summary":"","title":"Elliot Faugier"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/erwan-escudie/","section":"Authors","summary":"","title":"Erwan Escudie"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-04549409/","section":"Publications","summary":"","title":"Faster Optimal Coalition Structure Generation via Offline Coalition Selection and Graph-Based Search, Accepted paper"},{"content":" Scientific lead: YUN Bruno Is SyCoSMA the project lead? Yes Start date: 2024-01-01 End date: 2024-12-31 This project aims to develop the tools for an improved collaborative creation of argumentation graphs and their exploitation for assisted reasoning. Existing online argumentation platforms often depict a debate using a directed graph, offering a visual representation that enhances human grasp of the arguments and their interconnections. These platforms have various applications in domains such as education or e-democracy, where they facilitate broad public involvement in the development of laws. However, there are several problems preventing their adoption for assisted reasoning:\nInformation overload - As these platforms become increasingly prevalent, the volume of arguments and options within these systems continues to grow exponentially. This poses a significant hurdle for human analysts, who struggle to manually track changes, understand the various arguments and their relationships, and make sense of the vast amount of information; Lack of guidance - Users are not assisted in constructing their various arguments within a debate. This deficiency can result in irrationality, omission of critical information, and the establishment of relationships with incorrect polarity, particularly when confronted with complex arguments; Incomplete graphs - Users introduce new support or counter arguments into the system in reaction to a specific local argument. Typically, these new arguments may be redundant or engage with remote arguments that users have not taken into account, resulting in incomplete graphs. Consequently, the conventional argumentation semantics outlined in the literature become inapplicable; Asynchronous reasoning - Current research in real-time assisted reasoning within real-life debates using argumentation graphs remains limited. Existing approaches predominantly involve manual argument formalization or users selecting arguments from dropdown menus, interrupting the natural flow of the debate. This project will provide solutions for the aforementioned problems.\n","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/projects/13_hagarice/","section":"Axes","summary":"","title":"HAGARICE - Harnessing Argumentation Graphs in Augmented Reality for Immersive Co-Creation and Exploration"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/heike-baldauf-quilliatre/","section":"Authors","summary":"","title":"Heike Baldauf-Quilliatre"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-04720157/","section":"Publications","summary":"","title":"Human Compliance with Normative Principles in Argumentation: Effects of Naturalness Bias and Graphical Representation"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/jacques-saraydaryan/","section":"Authors","summary":"","title":"Jacques Saraydaryan"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/jean-christophe-routier/","section":"Authors","summary":"","title":"Jean-Christophe Routier"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/julien-perier-camby/","section":"Authors","summary":"","title":"Julien Perier-Camby"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/tags/knowledge-graphs/","section":"Tags","summary":"","title":"Knowledge Graphs"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-04849681/","section":"Publications","summary":"","title":"Le défi des données spontanées pour améliorer les interactions humain-robot : le projet Peppermint"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/tel-04846767/","section":"Publications","summary":"","title":"Learning spatial representations for single-task navigation and multi-task policies"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/lucien-tisserand/","section":"Authors","summary":"","title":"Lucien Tisserand"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/marie-lefevre/","section":"Authors","summary":"","title":"Marie Lefevre"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/marija-petrovi%C4%87/","section":"Authors","summary":"","title":"Marija Petrović"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-04711663/","section":"Publications","summary":"","title":"Multi-objective reinforcement learning: an ethical perspective"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-04711682/","section":"Publications","summary":"","title":"Multi-objective reinforcement learning: an ethical perspective"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/nathalie-guin/","section":"Authors","summary":"","title":"Nathalie Guin"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-04826283/","section":"Publications","summary":"","title":"Négociation pour la consommation adaptative d’allocation continue"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/olivier-l.-georgeon/","section":"Authors","summary":"","title":"Olivier L. Georgeon"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/olivier-simonin/","section":"Authors","summary":"","title":"Olivier Simonin"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/paul-robertson/","section":"Authors","summary":"","title":"Paul Robertson"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-04452246/","section":"Publications","summary":"","title":"Peppermint Conversational Framework (ppm-conv)"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/pierre-marza/","section":"Authors","summary":"","title":"Pierre Marza"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-04826205/","section":"Publications","summary":"","title":"Post-actes des Journées Francophones sur les Systèmes Multi-Agents (JFSMA 2023)"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-04691312/","section":"Publications","summary":"","title":"Reducing Intuitive-Physics Prediction Error through Playing"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/s%C3%A9bastien-konieczny/","section":"Authors","summary":"","title":"Sébastien Konieczny"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-05145796/","section":"Publications","summary":"","title":"Situated Self-guided Learning"},{"content":" Scientific lead: YUN Bruno Is SyCoSMA the project lead? Yes Start date: 2024-01-01 End date: 2026-01-01 In this project, we will instantiating a custom large-language model based on the recent Meta AI’s Llama 2 70B model and training it with both synthetic natural language arguments and arguments extracted from existing platforms. We will perform training through LoRA, a novel method to fine-tune LLMs that freezes the pre-trained model weights and adds trainable rank decomposition matrices to each layer of the Transformer architecture, which can capture task-specific features with fewer parameters. This synergy between argumentation graphs, knowledge graphs, and the LLM will be integrated in a decision-support system and evaluated in various decision scenarios.\n","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/projects/11_smarter/","section":"Axes","summary":"","title":"SMARTER - Structured Multi-sourced AI Reasoning for Enhanced Decision-Making "},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/tel-04858357/","section":"Publications","summary":"","title":"Structured representation learning and multimodal perception, towards more autonomous sensorimotor systems"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-04569375/","section":"Publications","summary":"","title":"Task-conditioned adaptation of visual features in multi-task policy learning"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-04748375/","section":"Publications","summary":"","title":"Unraveling the thread: understanding and addressing sequential failures in human-robot interaction"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-04093710/","section":"Publications","summary":"","title":"A Multiagent Path Search Algorithm for Large-Scale Coalition Structure Generation."},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/tel-04753578/","section":"Publications","summary":"","title":"Adaptive Learning en contexte parascolaire : modèles de recommandation et analyses des traces d'un déploiement industriel pour une amélioration continue du processus de personnalisation"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-04379426/","section":"Publications","summary":"","title":"Adaptive reinforcement learning of multi-agent ethically-aligned behaviours: the QSOM and QDSOM algorithms"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-04127943/","section":"Publications","summary":"","title":"AJAR: An Argumentation-based Judging Agents Framework for Ethical Reinforcement Learning"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/alexandre-albore/","section":"Authors","summary":"","title":"Alexandre Albore"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/alexandre-devillers/","section":"Authors","summary":"","title":"Alexandre Devillers"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-04241611/","section":"Publications","summary":"","title":"Algorithmic learning, a next step for AI. An application to arithmetic operations"},{"content":" Scientific lead: MATIGNON Laëtitia, PIERRE-LOUIS Olivier (ILM) Is SyCoSMA the project lead? Yes Start date: 2023-01-01 End date: 2023-12-31 This project is in collaboration with ILM laboratory, team MMCI.\nMorphological instabilities in out-of-equilibrium physical systems are numerous and diverse. One of the most well-known is fingering instability, which leads, for example, to the branched shapes seen in growing snowflakes. This type of instability belongs to a broad class observed across many systems, from hydrodynamics (viscous fingering) to combustion (flame front instability). For several decades, it has been extensively studied, revealing a wide variety of complex patterns that can emerge.\nThe core idea of our project is to control this instability using physical parameters, allowing us to navigate through the shape space toward a desired morphology. Existing theoretical models are currently unable to predict how to adjust physical parameters to reach a specific target shape. Therefore, we propose to explore model-free control of these morphologies through a reinforcement learning (RL) approach.\n","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/projects/2_almaric/","section":"Axes","summary":"","title":"almARIC - Controlling Growth Instabilities using Reinforcement Learning"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-04600410/","section":"Publications","summary":"","title":"An Affordance-Based Intersubjectivity Mechanism to Infer the Behaviour of Other Agents"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-04093961/","section":"Publications","summary":"","title":"An Information-Theoretic Perspective on Intrinsic Motivation in Reinforcement Learning: A Survey"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/anaelle-wilczynski/","section":"Authors","summary":"","title":"Anaelle Wilczynski"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/andrew-starkey/","section":"Authors","summary":"","title":"Andrew Starkey"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/anthony-baccuet/","section":"Authors","summary":"","title":"Anthony Baccuet"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/antoine-bouquin/","section":"Authors","summary":"","title":"Antoine Bouquin"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-04163030/","section":"Publications","summary":"","title":"Anytime Index-Based Search Method for Large-Scale Simultaneous Coalition Structure Generation and Assignment"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-04161020/","section":"Publications","summary":"","title":"Apprentissage de comportements éthiques multi-valeurs par combinaison d’agents juges symboliques et d’agents apprenants"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/arthur-aubret/","section":"Authors","summary":"","title":"Arthur Aubret"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/beno%C3%AEt-alcaraz/","section":"Authors","summary":"","title":"Benoît Alcaraz"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/benoit-le-blanc/","section":"Authors","summary":"","title":"Benoit Le Blanc"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/cassia-trojahn/","section":"Authors","summary":"","title":"Cassia Trojahn"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/catherine-roussey/","section":"Authors","summary":"","title":"Catherine Roussey"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-04164249/","section":"Publications","summary":"","title":"Champ neuronal et apprentissage profond de topologies pour la fusion multimodale"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/charlotte-laclau/","section":"Authors","summary":"","title":"Charlotte Laclau"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/christelle-launois/","section":"Authors","summary":"","title":"Christelle Launois"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/christopher-leturc/","section":"Authors","summary":"","title":"Christopher Leturc"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/cl%C3%A9ment-scheirlinck/","section":"Authors","summary":"","title":"Clément Scheirlinck"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-04092828/","section":"Publications","summary":"","title":"Comprendre les usages et effets d'un système de recommandations pédagogiques en contexte d'apprentissage non-formel"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-04310171/","section":"Publications","summary":"","title":"Conférence Nationale d’Intelligence Artificielle Année 2023"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-04198327/","section":"Publications","summary":"","title":"Cooperative Virtual Machine Placement"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-04166057/","section":"Publications","summary":"","title":"DisTop: Discovering a Topological representation to learn diverse and rewarding skills"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/dominique-longin/","section":"Authors","summary":"","title":"Dominique Longin"},{"content":" Scientific lead: MATIGNON Laëtitia, GUERIN-LASSOUS Isabelle Is SyCoSMA the project lead? Yes Start date: 2023-01-01 End date: 2024-12-31 This project, in collaboration with LIP (team Hownet) focuses on the quality of radio communications in drone fleets, which has a major impact on the overall flight performance of the fleet.\nUAV-based wireless networks can be deployed to provide a network coverage to users who have no or poor network connection. Unlike traditional model-based approaches that require predefined assumptions before UAV deployment, reinforcement learning (RL) offers a promising alternative but requires a realistic simulator for training the proposed strategies. None of the existing open-source simulators provide both realistic wireless communications while enabling the training, in a cluttered environment, of multi-UAVs movement strategies using RL. Thus, in this project, we developed a simulator where we focus on improving the modeling of the network access part, i.e., the communications between the UAVs and the users, by integrating signal propagation, physical rate adaptation and medium access sharing models. We evaluate the performance of a standard independent RL algorithm trained in our simulator across various use case scenarios, and compare the results obtained using different learning objectives. Results show that the classical formulation, commonly found in the literature and based on a a simplified wireless network model, underperforms in terms of communication quality.\n","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/projects/1_dronar/","section":"Axes","summary":"","title":"DRONAR - Reinforcement Learning for Drone Fleet Control with Communication Quality Awareness"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-04093716/","section":"Publications","summary":"","title":"Efficient Size-based Hybrid Algorithm for Optimal Coalition Structure Generation"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/%C3%A9lise-vareilles/","section":"Authors","summary":"","title":"Élise Vareilles"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/emmanuel-adam/","section":"Authors","summary":"","title":"Emmanuel Adam"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/emmanuel-rachelson/","section":"Authors","summary":"","title":"Emmanuel Rachelson"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-04184307/","section":"Publications","summary":"","title":"EquiMod: An Equivariance Module to Improve Visual Instance Discrimination"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-04335677/","section":"Publications","summary":"","title":"Ethical Smart Grid"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-04187666/","section":"Publications","summary":"","title":"Ethical Smart Grid: a Gym environment for learning ethical behaviours"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-04093715/","section":"Publications","summary":"","title":"Faster Optimal Coalition Structure Generation via Offline Coalition Selection and Graph-Based Search"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/fran%C3%A7ois-schwarzentruber/","section":"Authors","summary":"","title":"François Schwarzentruber"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/fran%C3%A7ois-suro/","section":"Authors","summary":"","title":"François Suro"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/hedwin-bonnavaud/","section":"Authors","summary":"","title":"Hedwin Bonnavaud"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-04317938/","section":"Publications","summary":"","title":"Inferring Attack Relations for Gradual Semantics"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/j%C3%A9r%C3%A9my-duval/","section":"Authors","summary":"","title":"Jérémy Duval"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-04164229/","section":"Publications","summary":"","title":"L'apprentissage algorithmique, une nouvelle étape pour l'IA. Une application aux opérations arithmétiques"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-05023726/","section":"Publications","summary":"","title":"Learning State Reachability as a Graph in Translation Invariant Goal-based Reinforcement Learning Tasks"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-04349804/","section":"Publications","summary":"","title":"Learning to identify and settle dilemmas through contextual user preferences"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/mathieu-buonomo/","section":"Authors","summary":"","title":"Mathieu Buonomo"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/tags/multi-agent-reinforcement-learning/","section":"Tags","summary":"","title":"Multi-Agent Reinforcement Learning"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/tags/multi-agent-systems/","section":"Tags","summary":"","title":"Multi-Agent Systems"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-04390240/","section":"Publications","summary":"","title":"Multi-Object Navigation with dynamically learned neural implicit representations"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/nir-oren/","section":"Authors","summary":"","title":"Nir Oren"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/olivier-boissier/","section":"Authors","summary":"","title":"Olivier Boissier"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-04317959/","section":"Publications","summary":"","title":"One-shot Learning for Task-oriented Grasping"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-04077910/","section":"Publications","summary":"","title":"Optimal Anytime Coalition Structure Generation Utilizing Compact Solution Space Representation"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-04077920/","section":"Publications","summary":"","title":"Parallel Index-based Search Algorithm for Coalition Structure Generation, student abstract"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/pascal-mei%C3%9Fner/","section":"Authors","summary":"","title":"Pascal Meißner"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/tags/physics/","section":"Tags","summary":"","title":"Physics"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/quenum-ghislain/","section":"Authors","summary":"","title":"Quenum Ghislain"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-04092808/","section":"Publications","summary":"","title":"Recommendation Model for an After-School E-learning Mobile Application"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/romaric-gaudel/","section":"Authors","summary":"","title":"Romaric Gaudel"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/salima-hassas/","section":"Authors","summary":"","title":"Salima Hassas"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/sandra-bringay/","section":"Authors","summary":"","title":"Sandra Bringay"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-04180331/","section":"Publications","summary":"","title":"Sequential annotations for naturally-occurring HRI: first insights"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/simon-forest/","section":"Authors","summary":"","title":"Simon Forest"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/tags/simulation/","section":"Tags","summary":"","title":"Simulation"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-04184291/","section":"Publications","summary":"","title":"Suréchantillonage Actif pour Modérer l'Apprentissage de Biais Visuels"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-04228330/","section":"Publications","summary":"","title":"The Enactive Inference Task in a Spatial Environment"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/thomas-guyet/","section":"Authors","summary":"","title":"Thomas Guyet"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/tags/uav-fleet/","section":"Tags","summary":"","title":"UAV Fleet"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-04209690/","section":"Publications","summary":"","title":"Un modèle pour la généricité des agents conversationnels vocaux multi-domaines"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-04092818/","section":"Publications","summary":"","title":"Understanding the Usages and Effects of a Recommendation System in a Non-formal Learning Context"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/valerija-holomjova/","section":"Authors","summary":"","title":"Valerija Holomjova"},{"content":" Scientific lead: LEFORT Mathieu Is SyCoSMA the project lead? Yes Start date: 2023-01-01 End date: 2023-12-01 Bien que l’avancée obtenue avec les auto-encodeurs variationnels ait eu un impact important dans le domaine de l’apprentissage de représentation, les méthodes contrastives de l\u0026rsquo;état de l\u0026rsquo;art ne capitalisent pas sur ses principes. Sachant que les auto-encodeurs variationnels permettent d\u0026rsquo;apprendre des caractéristiques décorrélées et que la décorrélation de caractéristiques a un impact positif sur l\u0026rsquo;effondrement dans les représentations nous souhaitons étudier dans ce projet si les représentations apprises par ces méthodes contrastives récentes peuvent bénéficier d\u0026rsquo;un aspect variationnel.\nPour cela, nous proposons dans un premier temps une modification des méthodes actuelles pour que les représentations produites soient des distributions et non des points dans l’espace de représentation. Puis dans un second temps, nous étudierons comment régulariser les distributions produites, comme c’est le cas dans les auto-encodeurs variationnels, afin d’éviter toute forme d\u0026rsquo;effondrement. Cette seconde étape nécessitera un travail d’adaptation étant donné que les méthodes d’apprentissage contrastif récentes utilisent un espace latent contraint à une hypersphère, là où les auto-encodeurs variationnels demandent un espace latent sans contrainte.\nAinsi avec ce travail nous espérons ainsi étudier comment l’apprentissage autosupervisé de représentation peut être combiné avec un aspect variationnel pour améliorer ses performances.\n","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/projects/33_vocal/","section":"Axes","summary":"","title":"VoCaL"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/tags/wireless-communication/","section":"Tags","summary":"","title":"Wireless Communication"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/zied-bouraoui/","section":"Authors","summary":"","title":"Zied Bouraoui"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/tags/classification/","section":"Tags","summary":"","title":"Classification"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/tags/explanability/","section":"Tags","summary":"","title":"Explanability"},{"content":" Scientific lead: SDIKA Michaël Is SyCoSMA the project lead? No Start date: 2022-03-10 End date: 2025-03-10 Nous souhaitons développer de nouvelles méthodes de pronostic du sepsis basées sur de l’apprentissage automatique, avec des prédictions explicables et compréhensibles par le personnel hospitalier, afin d’améliorer la prise en charge du patient, et notamment de réduire la durée de séjour hospitalier et la mortalité associée. Il s’agira ainsi de :\nProposer une structure originale d\u0026rsquo;apprentissage permettant d\u0026rsquo;améliorer l’explicabilité des modèles existants dans le domaine médical et la valider dans le contexte de prédiction du sepsis, Proposer une méthode de restitution des explications liées aux prédictions pertinentes pour le personnel hospitalier, Évaluer la valeur ajoutée de l’explicabilité et son impact sur le traitement du sepsis à l’hôpital. ","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/projects/23_previa_medical/","section":"Axes","summary":"","title":"PREVIA MEDICAL"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/tags/sepsis/","section":"Tags","summary":"","title":"Sepsis"},{"content":" Scientific lead: MATIGNON Laëtitia Is SyCoSMA the project lead? Yes Start date: 2022-01-10 End date: 2027-01-31 The project\u0026rsquo;s main goal is to enable the adaptive co-construction of Ethics in and for a long-lived intelligent system addressing these requirements. This raises 3 challenges:\nHuman-centric AI: how the system achieves its goal while following ethical principles and human values; Safe AI : how to ensure that the system operates within specified boundaries while being sufficiently autonomous to learn ethics and adapt in response to evolution of the context and objectives; Adaptive AI: AI systems should adapt both on a technical side (e.g., openness) and on a societal side (e.g., lifelong learning, shifting expectations and contemporary accepted moral values and norms). More information can be found here: https://projet.liris.cnrs.fr/acceler-ai/\n","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/projects/15_acceler_ia/","section":"Axes","summary":"","title":"ACCELER IA - Adaptive Co-Construction of Ethics for LifElong tRrustworthy AI"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-03600794/","section":"Publications","summary":"","title":"A dynamic neural field model of multimodal merging: application to the ventriloquist effect"},{"content":" Scientific lead: LEFORT Mathieu Is SyCoSMA the project lead? Yes Start date: 2022-01-01 End date: 2024-12-31 La fusion de données est une question cruciale en intelligence artificielle et en sciences cognitives. En effet, elle doit être résolue par tout agent, biologique ou artificiel, qui perçoit son environnement à travers plusieurs capteurs différenciés et complémentaires. Ce projet vise à mieux comprendre la perception chez l’humain afin d’améliorer les systèmes d’IA pour des agents plus autonomes. Il se positionne dans le cadre des théories sensori-motrices qui s’appuient sur un certain nombre d’indices provenant des neurosciences, de la psychologie, de la philosophie, … Plus particulièrement, nous considérons la théorie des contingences sensori-motrices qui propose un cadre unifié fondé sur 2 principes : 1- L’apprentissage de corrélations sensori-motrices qui pose en particulier la question de l’intégration multi-sensorielle, dès lors que l’on considère plusieurs capteurs, et même multimodale en considérant l’action. Une des questions centrales est de savoir quand et comment fusionner les informations disponibles de manière autonome en s’appuyant par exemple sur l’apprentissage passé des relations entre les différentes données. 2- La perception active qui formalise que l’être humain interagit avec son environnement non seulement pour atteindre ses objectifs ou maintenir son intégrité physique (action pragmatique), mais également afin d’obtenir plus d’informations sur son environnement (action épistémique), par exemple via les saccades oculaires. Cela nécessite à la fois de disposer d’un modèle prédictif de la relation à l’environnement (en conjonction avec l’apprentissage de représentation sensori-motrices) et de décider dans quelle(s) modalité(s) et dans quelle(s) zone(s) de l’espace aller chercher la prochaine information pertinente.\nCe projet va s’appuyer sur les travaux du projet AMPLIFIER et les poursuivre selon 3 directions : D1- La modélisation de la condition passive avec les champs neuronaux dynamiques (DNF) et en particulier l’étude de l’influence des hyperparamètres sur l’adéquation avec les données. Il s’agit de l’extension directe du travail fait dans mais appliqué aux données psychophysiques obtenues dans le projet AMPLIFIER. Cette modélisation computationnelle se fera conjointement avec une modélisation statistique fréquentiste et bayésienne des données, classiquement utilisée dans le cadre multi-sensoriel, réalisée au CRNL. En psychophysique, de nouvelles données seront collectées l’an prochain afin de tester les effets de la fiabilité et de la perception active dans une tâche de reconnaissance d’objets et non plus de localisation pure. Le cas échéant, elles seront intégrées dans ces modélisations et analysées afin de pouvoir comparer les différences entre une tâche de localisation et une de reconnaissance. D2- La comparaison des conditions passives et actives. Elle étendra la modélisation actuelle utilisant le paradigme des DNF, celui-ci étant compatible avec l’intégration d’anticipations sur la dynamique de l’environnement. Dans la littérature, des expériences de psychophysique montrent que dans le cas d’une modalité manquante, son influence diminue au fur et à mesure que sa précision ne peut être vérifiée dans l’environnement. Cela pourrait suggérer que le phénomène de capture visuelle serait plus prononcé dans le cas de la condition active (puisque l’information visuelle peut être explorée). Cependant les premières analyses de nos données expérimentales laissent entrevoir que le biais est dépendant de la cible de la saccade. La modélisation cherchera donc à expliquer ce phénomène, en particulier en s’intéressant aux différences inter-individuelles et inter-essais des participants, nos modèles permettant de les prendre en compte. D3- Le portage et l’adaptation des modèles produits par D1 et D2 dans un cadre robotique simulé, avec des stimuli auditifs et visuels simples, inspirés de ceux des expériences psychophysiques. L’objectif sera d’étudier plus en profondeur la dimension active de la perception dans l’estimation de la pertinence de chaque modalité. Cela permettra ainsi de proposer une solution de fusion multimodale originale et complémentaire des approches basées sur l’apprentissage profond.\n","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/projects/4_acme/","section":"Axes","summary":"","title":"ACME"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/alan-chauvin/","section":"Authors","summary":"","title":"Alan Chauvin"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/alisa-rieger/","section":"Authors","summary":"","title":"Alisa Rieger"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-03579876/","section":"Publications","summary":"","title":"An Abortion Based Search Method for Optimal Coalition Structure Generation"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-04168142/","section":"Publications","summary":"","title":"An information-theoretic perspective on intrinsic motivation in reinforcement learning"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/anael-belle/","section":"Authors","summary":"","title":"Anael Belle"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/anne-gwenn-bosser/","section":"Authors","summary":"","title":"Anne-Gwenn Bosser"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-03862226/","section":"Publications","summary":"","title":"Apprentissage pour la navigation robotique parmi les humains : vers des chemins s'adaptant aux dynamiques de l'environnement"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-03552075/","section":"Publications","summary":"","title":"Artificial Interactionism: Avoiding Isolating Perception From Cognition in AI"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/aymeric-h%C3%A9nard/","section":"Authors","summary":"","title":"Aymeric Hénard"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/tags/biological-evolution/","section":"Tags","summary":"","title":"Biological Evolution"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/c%C3%A9cile-daniel/","section":"Authors","summary":"","title":"Cécile Daniel"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/c%C3%A9line-teuli%C3%A8re/","section":"Authors","summary":"","title":"Céline Teulière"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-03693198/","section":"Publications","summary":"","title":"Combining Manifold Learning and Neural Field Dynamics for Multimodal Fusion"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-04168147/","section":"Publications","summary":"","title":"Compressed information is all you need: unifying intrinsic motivations and representation learning"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/tags/data-fusion/","section":"Tags","summary":"","title":"Data Fusion"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/tags/dynamical-systems/","section":"Tags","summary":"","title":"Dynamical Systems"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-04977898/","section":"Publications","summary":"","title":"Effect of sensory reliability and saccadic eye-movements on multisensory integration in the spatial ventriloquism paradigm"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/florence-dupin-de-saint-cyr/","section":"Authors","summary":"","title":"Florence Dupin De Saint-Cyr"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/florian-marchal/","section":"Authors","summary":"","title":"Florian Marchal"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-03828062/","section":"Publications","summary":"","title":"Identifying and localizing dynamic affordances to improve interactions with other agents"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-03879674/","section":"Publications","summary":"","title":"Interacting with Pepper: mutual learning of turn-taking practices in HRI (PepperMint)"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/jean-victor-autran/","section":"Authors","summary":"","title":"Jean-Victor Autran"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/jochen-triesch/","section":"Authors","summary":"","title":"Jochen Triesch"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/juan-r-vidal/","section":"Authors","summary":"","title":"Juan R Vidal"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/juliette-grosset/","section":"Authors","summary":"","title":"Juliette Grosset"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/tel-04098769/","section":"Publications","summary":"","title":"Large scale urban transport control for enhanced resilience : Analysis via complex networks, artificial intelligence and Big Data processing"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/tel-04107492/","section":"Publications","summary":"","title":"Learning behaviours aligned with moral values in a multi-agent system : guiding reinforcement learning with symbolic judgments"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/liana-ermakova/","section":"Authors","summary":"","title":"Liana Ermakova"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/marie-avillac/","section":"Authors","summary":"","title":"Marie Avillac"},{"content":" Scientific lead: LEFORT Mathieu Is SyCoSMA the project lead? Yes Start date: 2022-01-01 End date: 2024-12-31 Les oiseaux (plus de 10 000 espèces aujourd’hui) constituent la classe de Vertébrés Tétrapodes de loin la plus diversifiée. Ils figurent parmi les organismes les plus étudiés dans tous les domaines de la biologie. Des applications en IA de reconnaissance « automatique » se développent d’ailleurs pour aider à l’identification des espèces, sur critères d’apparence visuelle extérieure donc (e.g., Birdsnap ; Berg et al. 2014) (et d’autres sur les vocalisations ; Jancovic \u0026amp; Köküer 2012 ; Anand et al. 2021). Dans le même temps, des méthodes statistiques apparaissent pour modéliser l’apparence d’éléments du squelette d’ancêtres communs à des clades de vertébrés, à partir de morphologies actuelles et fossiles bien placées sur des arbres phylogénétiques maintenant bien robustes (ces arbres étant fondés sur la phylogénomique) (Coombs et al. 2020). Ces analyses morphologiques utilisent des landmarks et semi-landmarks (en nombre mais qui simplifient et fractionnent malgré tout la morphologie de façon arbitraire). Or, jusqu’ici rien n’a été tenté sur la modélisation d’ancêtres de clades entiers d’oiseaux en termes d’apparence visuelle extérieure (et encore moins acoustique). Cela comprend le plumage avec ses patterns et ses couleurs, et les parties nues, ainsi que la silhouette générale (au posé ou en vol) qui délimite l’oiseau. Dans ce projet nous allons concevoir et mettre en œuvre des méthodes d\u0026rsquo;apprentissage avec des réseaux de neurones génératifs pour modéliser l\u0026rsquo;évolution des espèces d\u0026rsquo;oiseaux en se basant sur leur apparence. Un premier travail consisterait en la constitution d\u0026rsquo;une base d\u0026rsquo;apprentissage adaptée à l\u0026rsquo;apprentissage d\u0026rsquo;un tel modèle avec des images extraites de référentiels comme Birds of the World (dessins), le choix de codage et normalisation de ces données etc., ainsi que l\u0026rsquo;architecture du modèle, l\u0026rsquo;algorithme d\u0026rsquo;apprentissage et la fonction à optimiser. Une première approche serait alors d\u0026rsquo;utiliser un VAE convolutif pour apprendre une représentation plus compacte des images, et ensuite d\u0026rsquo;utiliser cette représentation latente en entrée des modèles existants utilisés par des biologistes basés sur l\u0026rsquo;optimisation sous contraintes par des connaissances du métier (phylogénétiques). Une deuxième étape serait d\u0026rsquo;intégrer ces contraintes directement dans l\u0026rsquo;apprentissage du modèle, soit par la fonction de coût (ou la régularisation) soit par l\u0026rsquo;algorithme d\u0026rsquo;apprentissage (par exemple la manière dont les données sont organisées, étiquetées et/ou présentées). Un défi est alors de concevoir un modèle qui apprend une représentation continue qui peut avoir un sens pour les experts du domaine (biologique, paléontologique etc.) intégrant à la fois l\u0026rsquo;aspect temporel (évolution) et proximité génétique (ancêtres communs etc.). Une approche serait d\u0026rsquo;utiliser des méthodes faiblement supervisées, par exemple l\u0026rsquo;apprentissage contrastif, l\u0026rsquo;apprentissage d\u0026rsquo;une \u0026ldquo;métrique\u0026rdquo; de similarité en présentant des paires ou triplet d\u0026rsquo;oiseaux appartenant à la même espèce ou non ou la création de modèles séparés pour différentes modalités ou espaces sémantiques (espèces, caractéristiques) et les fusionner plus en aval dans le traitement. Une des questions étudiées sera de savoir comment représenter certaines discontinuités, les relations hiérarchiques et les propriétés topologiques dans ce \u0026ldquo;manifold\u0026rdquo; (variété géométrique) appris par le réseaux de neurones pour explicitement intégrer certaines contraintes. Une fois le modèle appris, il pourrait être utilisé pour générer un \u0026ldquo;portrait-robot\u0026rdquo; d\u0026rsquo;un oiseau à partir d\u0026rsquo;un point de la représentation latente (ayant certaines propriétés souhaitées, par exemple un ancêtre commun de deux groupes), et ensuite donner lieu à des questionnements ou alors des nouvelles connaissances d\u0026rsquo;experts du domaine. Certaines connaissances pourraient également provenir de la représentation interne apprise automatiquement. Par exemple, la proximité/l\u0026rsquo;éloignement de points dans l\u0026rsquo;espace, la relation entre deux ou plusieurs exemples etc.\n","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/projects/0_merle/","section":"Axes","summary":"","title":"MERLE"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/nathalie-guyader/","section":"Authors","summary":"","title":"Nathalie Guyader"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/nicolas-p%C3%A9pin-hermann/","section":"Authors","summary":"","title":"Nicolas Pépin-Hermann"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-03349170/","section":"Publications","summary":"","title":"Novel Decision-Making Strategy for Connected and Autonomous Vehicles in Highway On-ramp Merging"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-03830080/","section":"Publications","summary":"","title":"PICS: Parallel Index-based Search Algorithm for Coalition Structure Generation"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/pierre-de-loor/","section":"Authors","summary":"","title":"Pierre De Loor"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-03795272/","section":"Publications","summary":"","title":"Poetic or Humorous Text Generation: Jam Event at PFIA2022"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/tags/psychophysics-modeling/","section":"Tags","summary":"","title":"Psychophysics Modeling"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/rebiha-bacha/","section":"Authors","summary":"","title":"Rebiha Bacha"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-03846074/","section":"Publications","summary":"","title":"Recommandation de ressources éducatives en contexte d’apprentissage parascolaire sur application mobile"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/tags/robotic/","section":"Tags","summary":"","title":"Robotic"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-04060326/","section":"Publications","summary":"","title":"Simultaneous Localization and Active Phenomenon Inference (SLAPI)"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/tags/social-navigation/","section":"Tags","summary":"","title":"Social Navigation"},{"content":" Scientific lead: MATIGNON Laëtitia, SARAYDARYAN Jacques (CITI) Is SyCoSMA the project lead? Yes Start date: 2022-01-01 End date: 2023-12-31 This project is in collaboration with CITI laboratory, CHROMA team.\nIn this project, we developed a novel reinforcement learning-based approach for socially aware multi-agent navigation.\n","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/projects/3_socianav/","section":"Axes","summary":"","title":"SociaNav - Multi-Agent Reinforcement Learning-Based Robotic Navigation Among Humans"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-03830082/","section":"Publications","summary":"","title":"Subspace-Focused Search Method for Optimal Coalition Structure Generation"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-03940658/","section":"Publications","summary":"","title":"Teaching Agents how to Map: Spatial Reasoning for Multi-Object Navigation"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-03815546/","section":"Publications","summary":"","title":"The Impact of Action in Visual Representation Learning"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/titouan-knockaert/","section":"Authors","summary":"","title":"Titouan Knockaert"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/tel-04069497/","section":"Publications","summary":"","title":"Topologies en perception multimodale neuro-inspirée : apprentissage, fusion, prise de décision"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-03765558/","section":"Publications","summary":"","title":"Towards Considering Explicit Sensitivity to Augmentation in Visual Instance Discrimination Tasks"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-04977890/","section":"Publications","summary":"","title":"Unimodal Weights in Bimodal Perception: Influence of intermodal differences in stimulus reliability and task induced saccades on crossmodal bias"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/valentin-chaffraix/","section":"Authors","summary":"","title":"Valentin Chaffraix"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/victor-charpenay/","section":"Authors","summary":"","title":"Victor Charpenay"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/yohan-michelland/","section":"Authors","summary":"","title":"Yohan Michelland"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/tags/human-robot-interaction/","section":"Tags","summary":"","title":"Human-Robot Interaction"},{"content":" Scientific lead: ARMETTA Is SyCoSMA the project lead? Yes Start date: 2021-09-01 End date: 2025-03-31 PepperMint (Interacting with Pepper: mutual learning of turn-taking practices in HRI) is funded by the ASLAN Labex. It proposes an explorative study of embodied turn-taking practices in task-related Human-Robot Interaction (HRI) to improve the social abilities of robots and make HRI more natural to humans. The project initiates a cooperation between researchers in AI (Artificial Intelligence) (LIRIS) and CA (Conversation Analysis) (ICAR and GenZ Oulu - Finland). It investigates if and how CA findings on natural occurring interaction can be used to develop innovative and effective AI models for HRI. The project is grounded in a detailed multimodal analysis of turn-taking in naturally occurring HRI, putting forward the emergence of turn allocation as complex sequential and multimodal practices.\nProject website\n","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/projects/26_peppermint/","section":"Axes","summary":"","title":"PepperMint"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/tags/turn-taking/","section":"Tags","summary":"","title":"Turn-Taking"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/emse-03318195/","section":"Publications","summary":"","title":"A Multi-Agent Approach to Combine Reasoning and Learning for an Ethical Behavior"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/alain-mille/","section":"Authors","summary":"","title":"Alain Mille"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/am%C3%A9lie-cordier/","section":"Authors","summary":"","title":"Amélie Cordier"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-03455635/","section":"Publications","summary":"","title":"Apprentissage séquentiel de compétences via la motivation intrinsèque et l'apprentissage par renforcement"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/emse-03278353/","section":"Publications","summary":"","title":"Approche multi-agent combinant raisonnement et apprentissage pour un comportement éthique"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/ashish-kumar-maurya/","section":"Authors","summary":"","title":"Ashish Kumar Maurya"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-03091643/","section":"Publications","summary":"","title":"BOSS: A Bi-directional Search Technique for Optimal Coalition Structure Generation with Minimal Overlapping, Student Abstract"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-03619574/","section":"Publications","summary":"","title":"Coalition Formation with Multiple Alternatives of Interdependent Tasks"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-03378938/","section":"Publications","summary":"","title":"Code-based Algorithm for Coalition Structure Generation"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-03352684/","section":"Publications","summary":"","title":"DisTop: Discovering a Topological representation to learn diverse and rewarding skills"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-03106286/","section":"Publications","summary":"","title":"Explanation for Humans, for Machines, for Human-Machine Interactions?"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-03091652/","section":"Publications","summary":"","title":"FACS: Fast code-based Algorithm for Coalition Structure Generation, Student Abstract, Selected for presentation"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/florian-pereyron/","section":"Authors","summary":"","title":"Florian Pereyron"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-02532976/","section":"Publications","summary":"","title":"FRLLE: A Failure Rate and Load based Leader Election Algorithm for a Bidirectional Ring in Distributed Systems"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/guillaume-bono/","section":"Authors","summary":"","title":"Guillaume Bono"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/hanane-ariouat/","section":"Authors","summary":"","title":"Hanane Ariouat"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-02569235/","section":"Publications","summary":"","title":"Improving Coalition Structure Search with an Imperfect Algorithm : Analysis and Evaluation Results"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/jilles-dibangoye/","section":"Authors","summary":"","title":"Jilles Dibangoye"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-03199996/","section":"Publications","summary":"","title":"Lea-TN: Leader Election Algorithm Considering Node and Link Failures in a Torus Network"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/tel-03684227/","section":"Publications","summary":"","title":"Learning increasingly complex skills through deep reinforcement learning using intrinsic motivation"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-03091646/","section":"Publications","summary":"","title":"Leveraging on Deep Reinforcement Learning for Autonomous Safe Decision-Making in Highway On-ramp Merging, Student Abstract"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-03465146/","section":"Publications","summary":"","title":"Novelty detection for unsupervised continual learning in image sequences"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/onn-shehory/","section":"Authors","summary":"","title":"Onn Shehory"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/ruiqi-dai/","section":"Authors","summary":"","title":"Ruiqi Dai"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/sarvapali-ramchurn/","section":"Authors","summary":"","title":"Sarvapali Ramchurn"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-03465149/","section":"Publications","summary":"","title":"Self-supervised continual learning for object recognition in image sequences"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-02909132/","section":"Publications","summary":"","title":"Solving Multi-Agent Routing Problems Using Deep Attention Mechanisms"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-03287742/","section":"Publications","summary":"","title":"Système de recommandation de ressources pédagogiques pour un apprentissage sur application mobile parascolaire"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/tags/intelligent-tutoring-systems/","section":"Tags","summary":"","title":"Intelligent Tutoring Systems"},{"content":" Scientific lead: LEFEVRE Marie Is SyCoSMA the project lead? No Start date: End date: Les élèves qui utilisent des outils technologiques éducatifs parascolaires comme l’application mobile Nomad Education recherchent un complément à la formation qu’ils reçoivent en classe. En effet, les conditions d’enseignement en classe dans l’Éducation Nationale (créneaux alloués à chaque matière dans l’emploi du temps, effectif des groupes) ne permettent pas toujours de s’adapter aux besoins des élèves. Certains d’entre eux téléchargent donc l’application Nomad Education afin de retravailler certains sujets de leur programme scolaire avec une approche pédagogique différente. L’utilisation de l’application est volontaire, parascolaire et non-encadrée. L’objectif de mes travaux est d’intégrer à l’application mobile existante un système de recommandations de contenus pédagogiques, pour accompagner l’apprenant de manière personnalisée dans son utilisation de l’application. L’enjeu est de permettre à l’élève de s’abstraire de son niveau scolaire en travaillant des ressources en fonction de ses besoins et acquis et de faire des liens entre les différents chapitres qu’il pourra rencontrer au cours de son apprentissage. On souhaite pour cela un système qui soit à la fois adaptatif et intelligent.\nLes contributions proposées s’attachent à apporter des réponses à la problématique suivante:\nComment proposer un système de recommandations pédagogiques adapté à une utilisation en contexte parascolaire non encadré en exploitant exclusivement le contenu d’une application mobile et l’activité des apprenants au sein de cette application ?\nL\u0026rsquo;identification des besoins d\u0026rsquo;apprentissage de chaque utilisateur est une tâche complexe. Antérieurement à ce doctorat, Nomad Education a entrepris l\u0026rsquo;application de la théorie de la réponse aux items (Item Response Theory, IRT) à ses contenus dans l\u0026rsquo;optique de l\u0026rsquo;intégrer au système de recommandations. L\u0026rsquo;intégration de l\u0026rsquo;IRT pour répondre à cette problématique amène une première question de recherche:\n(QR-1): Peut-on utiliser la théorie de la réponse aux items (IRT) afin d\u0026rsquo;identifier les besoins d\u0026rsquo;apprentissage des apprenants en s’appuyant sur leurs réponses aux quiz ?\nDans un contexte parascolaire nous n\u0026rsquo;avons qu\u0026rsquo;une vue partielle de l\u0026rsquo;apprentissage de l\u0026rsquo;apprenant. L\u0026rsquo;utilisation de l\u0026rsquo;application étant volontaire, le système de recommandation doit répondre au besoin ponctuel de l’apprenant et s\u0026rsquo;intégrer dans son utilisation de l\u0026rsquo;application mobile. La deuxième question de recherche est la suivante:\n(QR-2): Comment proposer des activités pédagogiquement pertinentes et adaptées aux souhaits d\u0026rsquo;étude ponctuels de l\u0026rsquo;apprenant, sans programme pédagogique préétabli, en contexte de micro-learning?\nPour répondre à ma problématique, j\u0026rsquo;ai proposé un modèle permettant de mettre en place un système de recommandation pour une application mobile utilisée en contexte parascolaire. Ce modèle est l\u0026rsquo;objet de ma première contribution.\nAfin d\u0026rsquo;étudier la QR-1, j\u0026rsquo;utilise la théorie de la réponse aux items pour identifier chez les apprenants des groupes de niveaux. Je fais l\u0026rsquo;hypothèse qu\u0026rsquo;à chaque groupe de niveau correspondent des besoins de recommandation, pour lesquels je définis des stratégies de personnalisation.\nPour répondre à la QR-2, je propose un score de recommandation s’appuyant sur 3 composantes (pédagogique, historique et nouveauté) afin de classer les ressources sélectionnées par la stratégie de recommandation préalablement identifiée.\nAssocié à ce modèle de recommandation combinant IRT et score de recommandation, je propose un processus pour l\u0026rsquo;implémenter dans le cadre de l\u0026rsquo;application Nomad Education. Cette première contribution a été évaluée avec des enseignants afin de valider l\u0026rsquo;intérêt pédagogique des recommandations, puis par l\u0026rsquo;analyse de l\u0026rsquo;utilisation des recommandations par les apprenants en contexte écologique.\nDans ma deuxième contribution, je m\u0026rsquo;appuie sur les analyses faites du modèle proposé dans la première contribution afin d\u0026rsquo;en proposer des alternatives. Je reviens sur la QR-1 en étudiant l\u0026rsquo;impact des différents paramètres d\u0026rsquo;exploitation de l\u0026rsquo;IRT dans le modèle initial. Je propose des variations permettant de modifier les critères de l\u0026rsquo;IRT pour créer les groupes de niveau, requestionnant la stratégie de personnalisation associée à chaque niveau IRT et laissant à l\u0026rsquo;apprenant la possibilité de choisir lui-même la stratégie de personnalisation en dehors de celle qui lui a été assignée à partir de l\u0026rsquo;IRT. Je reviens également sur l\u0026rsquo;interface de présentation des recommandations mise en place dans la première contribution pour étudier si l\u0026rsquo;affichage du niveau scolaire des contenus recommandés peut impacter le suivi des recommandations.\nDans ma troisième contribution, je pars de l\u0026rsquo;analyse de l\u0026rsquo;utilisation du système de recommandation par les apprenants pour venir compléter la réponse apportée à la QR-2 dans la première contribution. L\u0026rsquo;analyse des traces ayant mis en évidence que certaines ressources sont davantage plébiscitées, je propose un processus de découverte des connaissances, permettant d\u0026rsquo;extraire des traces d\u0026rsquo;utilisation des informations pertinentes pouvant enrichir la base de connaissances sur laquelle le modèle de recommandation proposé s\u0026rsquo;appuie. Cette découverte de connaissances repose d\u0026rsquo;une part sur les notions abordées dans les recommandations et d\u0026rsquo;autre part sur l\u0026rsquo;enchaînement des activités au sein de l\u0026rsquo;application.\n","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/projects/28_nomad/","section":"Axes","summary":"","title":"Nomad Education"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/tags/recommander-systems/","section":"Tags","summary":"","title":"Recommander Systems"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-03104997/","section":"Publications","summary":"","title":"A Constructivist Approach and Tool for Autonomous Agent Bottom-up Sequential Learning"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/abdelkamel-tari/","section":"Authors","summary":"","title":"Abdelkamel Tari"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-04131714/","section":"Publications","summary":"","title":"Acquisition of Autonomy in Biotechnology and Artificial Intelligence"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/alessandro-farinelli/","section":"Authors","summary":"","title":"Alessandro Farinelli"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-04077921/","section":"Publications","summary":"","title":"An Extended Multi-agent Coalitions Mechanism with Constraints"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/anouer-bennajeh/","section":"Authors","summary":"","title":"Anouer Bennajeh"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/antoine-gr%C3%A9a/","section":"Authors","summary":"","title":"Antoine Gréa"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-03012127/","section":"Publications","summary":"","title":"Apprentissage adaptatif de comportements éthiques"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/tel-03406086/","section":"Publications","summary":"","title":"Architecture cognitive constructiviste : un modèle pour concevoir un agent automotivé capable de faire du sens et de construire des connaissances de l'environnement"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/bastien-nollet/","section":"Authors","summary":"","title":"Bastien Nollet"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/b%C3%A9atrice-de-montera/","section":"Authors","summary":"","title":"Béatrice De Montera"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/benoit-vuillemin/","section":"Authors","summary":"","title":"Benoit Vuillemin"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-02484960/","section":"Publications","summary":"","title":"Coalition formation with dynamically changing externalities"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-02975529/","section":"Publications","summary":"","title":"Continuous Learning of Action and State Spaces (CLASS)"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/cyrus-shaoul/","section":"Authors","summary":"","title":"Cyrus Shaoul"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/tel-03347713/","section":"Publications","summary":"","title":"Decentralised congestion control mechanisms for autonomous vehicles using distributed constraints optimisation"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-02500536/","section":"Publications","summary":"","title":"Decentralised Control of Intelligent Devices: A Healthcare Facility Study"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-02500538/","section":"Publications","summary":"","title":"Decentralised Multi-Intersection Congestion Control for Connected Autonomous Vehicles"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-02500542/","section":"Publications","summary":"","title":"Distributed multi-issue multi-lateral negotiation using a divide and rule approach"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-02902567/","section":"Publications","summary":"","title":"ELSIM: End-to-end learning of reusable skills through intrinsic motivation"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-02902570/","section":"Publications","summary":"","title":"ELSIM: End-to-end learning of reusable skills through intrinsic motivation"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-02902573/","section":"Publications","summary":"","title":"ELSIM: End-to-end learning of reusable skills through intrinsic motivation"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-02930211/","section":"Publications","summary":"","title":"Generating Natural Behaviors using Constructivist Algorithms"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/henry-minsky/","section":"Authors","summary":"","title":"Henry Minsky"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/huan-vu/","section":"Authors","summary":"","title":"Huan Vu"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-02975545/","section":"Publications","summary":"","title":"IWSSL Introduction to this volume"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/jianyong-xue/","section":"Authors","summary":"","title":"Jianyong Xue"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/lamjed-ben-said/","section":"Authors","summary":"","title":"Lamjed Ben Said"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-02929738/","section":"Publications","summary":"","title":"Learning Arithmetic Operations With A Multistep Deep Learning"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/tel-03288697/","section":"Publications","summary":"","title":"Méta-langage endomorphe et planification abstraite pour la reconnaissance des intentions en temps réel"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/milan-minsky/","section":"Authors","summary":"","title":"Milan Minsky"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-02595335/","section":"Publications","summary":"","title":"Multi-agent cooperation for an active perception based on driving behavior: Application in a car-following behavior"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/ndeye-arame-diago/","section":"Authors","summary":"","title":"Ndeye Arame Diago"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-02357974/","section":"Publications","summary":"","title":"New Off-board Solution for Predicting Vehicles' Intentions in the Highway On-Ramp using Probabilistic Classifiers, Long Abstract"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-02979148/","section":"Publications","summary":"","title":"Novel Off-board Solution for Predicting Drivers’ Intentions using Probabilistic Classifiers"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-02357971/","section":"Publications","summary":"","title":"ODSS: Efficient Hybridization for Optimal Coalition Structure Generation"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/philippe-gagnon/","section":"Authors","summary":"","title":"Philippe Gagnon"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/tel-02934428/","section":"Publications","summary":"","title":"Prediction rule mining in an Ambient Intelligence context"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/sacha-lhopital/","section":"Authors","summary":"","title":"Sacha Lhopital"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/slim-bechikh/","section":"Authors","summary":"","title":"Slim Bechikh"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-03352469/","section":"Publications","summary":"","title":"Une perspective historique sur l'IA explicable Document préparatoire à un tutorial AFIA juillet 2020"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/vincent-thavonekham/","section":"Authors","summary":"","title":"Vincent Thavonekham"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/tel-02519746/","section":"Publications","summary":"","title":"A bottom-up approach to energy management in a Smart Grid : adaptive and reactive model using a decentralized architecture for a generic user-centered system enabling large scale deployments"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-01910489/","section":"Publications","summary":"","title":"A Dynamic Bayesian Network Based Merge Mechanism for Autonomous Vehicles, Long Abstract"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-02098831/","section":"Publications","summary":"","title":"A preliminary study to compare deep learning with rule-based approaches for citation classification"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/alexander-riegler/","section":"Authors","summary":"","title":"Alexander Riegler"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/alexandre-galdeano/","section":"Authors","summary":"","title":"Alexandre Galdeano"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-02266791/","section":"Publications","summary":"","title":"An Effective Dynamic Programming Algorithm for Optimal Coalition Structure Generation"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-01910491/","section":"Publications","summary":"","title":"An Imperfect Algorithm for Coalition Structure Generation, Long Abstract"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-02158807/","section":"Publications","summary":"","title":"An Improved Algorithm for Optimal Coalition Structure Generation"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/andr%C3%A9-fabbri/","section":"Authors","summary":"","title":"André Fabbri"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/antoine-grea/","section":"Authors","summary":"","title":"Antoine Grea"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/tags/automatic/","section":"Tags","summary":"","title":"Automatic"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-02302101/","section":"Publications","summary":"","title":"Bi-level decision-making modeling for an autonomous driver agent: Application in the car-following driving behavior"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-03694078/","section":"Publications","summary":"","title":"Biological constraints in neural field models of sensor fusion"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-02270223/","section":"Publications","summary":"","title":"CASH Only: Constitutive Autonomy through Motorsensory Self-Programming"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/tags/control-theory/","section":"Tags","summary":"","title":"Control Theory"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-01969591/","section":"Publications","summary":"","title":"DACYCLEM: A decentralized algorithm for maximizing coverage and lifetime in a mobile wireless sensor network"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-03498605/","section":"Publications","summary":"","title":"De la signification éthique des limites"},{"content":" Scientific lead: MATIGNON Laëtitia for LIRIS, Madiha Nadri (LAGEPP) for the project Is SyCoSMA the project lead? No Start date: 2019-01-01 End date: 2024-12-31 Recent progress in machine learning (ML) is often attributed to methodological advances, as well as massive amounts of training data and compute. We propose to extend these advances to sequential decision-making of multiple agents for planning and control. Our focus is on learning realistic behaviors that require long-term planning and robust, stable control in the short term. Methodologically, this project is at cross roads of machine learning and control theory. These two disciplines have a long and rich history of interactions between them and their overlap is becoming more and more evident. The project proposes fundamental contributions: adding stability to the algorithms of reinforcement learning; data driven methods for robust control; hybrid ML / CT methods for multi-horizon control and planning; decentralized control. The methodological contributions of this fundamental IA project will be applied to the robust control of UAV fleets.\nMore information here: https://projet.liris.cnrs.fr/delicio/\n","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/projects/27_delicio/","section":"Axes","summary":"","title":"DeLiCio - Data and Priors, Learning and Control"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-02458953/","section":"Publications","summary":"","title":"Determining Citation Blocks using End-to-end Neural Coreference Resolution Model for Citation Context Analysis"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-01990340/","section":"Publications","summary":"","title":"Dynamic and intelligent control of autonomous vehicles for highway on-ramp merge,"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-02272091/","section":"Publications","summary":"","title":"Étude de la motivation intrinsèque en apprentissage par renforcement"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/fr%C3%A9d%C3%A9ric-guinand/","section":"Authors","summary":"","title":"Frédéric Guinand"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-02097488/","section":"Publications","summary":"","title":"HEART: Using Abstract Plans as a Guarantee of Downward Refinement in Decompositional Planning"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/iana-atanassova/","section":"Authors","summary":"","title":"Iana Atanassova"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-02271586/","section":"Publications","summary":"","title":"Identifying the Conceptual Space of Citation Contexts using Coreferences"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/jean-matthieu-etancelin/","section":"Authors","summary":"","title":"Jean-Matthieu Etancelin"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/laurianne-charrier/","section":"Authors","summary":"","title":"Laurianne Charrier"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-02275636/","section":"Publications","summary":"","title":"Leveraging Symmetric Relations for Approximation Coalition Structure Generation, To appear"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/lionel-delphin-poulat/","section":"Authors","summary":"","title":"Lionel Delphin-Poulat"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/tags/machine-learning/","section":"Tags","summary":"","title":"Machine Learning"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/marc-bertin/","section":"Authors","summary":"","title":"Marc Bertin"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/martin-rosalie/","section":"Authors","summary":"","title":"Martin Rosalie"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-02013653/","section":"Publications","summary":"","title":"Multi-Agent Learning and Coordination with Clustered Deep Q-Network"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/pierre-jonin/","section":"Authors","summary":"","title":"Pierre Jonin"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/rozenn-nicol/","section":"Authors","summary":"","title":"Rozenn Nicol"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/simon-pageaud/","section":"Authors","summary":"","title":"Simon Pageaud"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/tel-02475605/","section":"Publications","summary":"","title":"SmartGov : architecture générique pour la co-construction de politiques urbaines basée sur l'apprentissage par renforcement multi-agent"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-04077844/","section":"Publications","summary":"","title":"The RoPE Scale: a Measure of How Empathic a Robot is Perceived"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-02190737/","section":"Publications","summary":"","title":"TSRuleGrowth : Extraction de règles de prédiction semi-ordonnées à partir d'une série temporelle d'éléments discrets, application dans un contexte d'intelligence ambiante"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-02305993/","section":"Publications","summary":"","title":"TSRuleGrowth: Mining partially-ordered prediction rules from a time series of discrete elements, application to a context of ambient intelligence"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/vassilissa-lehoux-lebacque/","section":"Authors","summary":"","title":"Vassilissa Lehoux-Lebacque"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/v%C3%A9ronique-deslandres/","section":"Authors","summary":"","title":"Véronique Deslandres"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/victor-lequay/","section":"Authors","summary":"","title":"Victor Lequay"},{"content":" Scientific lead: LEFORT Mathieu Is SyCoSMA the project lead? Yes Start date: 2018-03-20 End date: 2023-05-15 Scientific partners: LIMOS, UCLy Industrial partner: Ubiant The Ethics.AI project aims to propose an intelligent, responsible system, interacting with humans, capable of autonomously developing an embodied ethic within the framework of this interaction, in order to adopt behaviours adapted to situations and in conformity with human ethical behaviour patterns.\nSee the project website\n","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/projects/20_ethics_ai/","section":"Axes","summary":"","title":"Ethics.AI"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/tags/smart-grid/","section":"Tags","summary":"","title":"Smart Grid"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/tags/socio-technical-systems/","section":"Tags","summary":"","title":"Socio-Technical Systems"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-01772181/","section":"Publications","summary":"","title":"A Decentralised Approach to Intersection Traffic Management"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-01791523/","section":"Publications","summary":"","title":"A fuzzy logic-based anticipation car-following model"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/adrien-techer/","section":"Authors","summary":"","title":"Adrien Techer"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/aline-parreau/","section":"Authors","summary":"","title":"Aline Parreau"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/alix-gonnot/","section":"Authors","summary":"","title":"Alix Gonnot"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-01897042/","section":"Publications","summary":"","title":"Anticipation model based on a modified fuzzy logic approach"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-01885654/","section":"Publications","summary":"","title":"Apprentissage constructiviste à base de systèmes multiagents. Une application au problème complexe de la régulation coopérative du trafic"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-01866197/","section":"Publications","summary":"","title":"Causality Reconstruction by an Autonomous Agent"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/cl%C3%A9ment-cottet/","section":"Authors","summary":"","title":"Clément Cottet"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-01832613/","section":"Publications","summary":"","title":"Context Aware Robot Architecture, Application to the RoboCup@Home Challenge"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-01847775/","section":"Publications","summary":"","title":"Cooperative Multi-Agent Policy Gradient"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-01852233/","section":"Publications","summary":"","title":"Developmental Learning for Social Robots in Real-World Interactions"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-01887075/","section":"Publications","summary":"","title":"Empathy Display Influence on Human-Robot Interactions: a Pilot Study"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/eric-duch%C3%AAne/","section":"Authors","summary":"","title":"Eric Duchêne"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/eric-lombardi/","section":"Authors","summary":"","title":"Eric Lombardi"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/eric-sopena/","section":"Authors","summary":"","title":"Eric Sopena"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/fabrice-jumel/","section":"Authors","summary":"","title":"Fabrice Jumel"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-01834701/","section":"Publications","summary":"","title":"HEART: HiErarchical Abstraction for Real-Time Partial Order Causal Link Planning"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-01878646/","section":"Publications","summary":"","title":"How explainable plans can make planning faster"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-01170310/","section":"Publications","summary":"","title":"Ice sliding games"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-01894621/","section":"Publications","summary":"","title":"Influence of eye-movements on multisensory stimulus localization: experiments, models and robotics applications"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/jilles-s-dibangoye/","section":"Authors","summary":"","title":"Jilles S Dibangoye"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/jilles-steeve-dibangoye/","section":"Authors","summary":"","title":"Jilles Steeve Dibangoye"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/jos%C3%A9-ghislain-quenum/","section":"Authors","summary":"","title":"José Ghislain Quenum"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/julien-moncel/","section":"Authors","summary":"","title":"Julien Moncel"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-01738132/","section":"Publications","summary":"","title":"Mastering the Laws of Feedback Contingencies Is Essential to Constructivist Artificial Agents"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/maxime-gu%C3%A9riau/","section":"Authors","summary":"","title":"Maxime Guériau"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/mbaye-s%C3%A8ne/","section":"Authors","summary":"","title":"Mbaye Sène"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/tel-02027250/","section":"Publications","summary":"","title":"Mécanismes de négociation multilatérale pour la prise de décision collective"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-01774799/","section":"Publications","summary":"","title":"Multi-Robot Simultaneous Coverage and Mapping of Complex Scene"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-01726120/","section":"Publications","summary":"","title":"Multi-Robot Simultaneous Coverage and Mapping of Complex Scene - Comparison of Different Strategies"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-01793456/","section":"Publications","summary":"","title":"Négociation multilatérale pour la prise de décision collective"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/nour-eddin-el-faouzi/","section":"Authors","summary":"","title":"Nour Eddin El Faouzi"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-01821677/","section":"Publications","summary":"","title":"On the Study of Cooperative Multi-Agent Policy Gradient"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/paul-dorbec/","section":"Authors","summary":"","title":"Paul Dorbec"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/raphael-leber/","section":"Authors","summary":"","title":"Raphael Leber"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/romain-billot/","section":"Authors","summary":"","title":"Romain Billot"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-01847773/","section":"Publications","summary":"","title":"SULFR: Simulation of Urban Logistic For Reinforcement"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-01840852/","section":"Publications","summary":"","title":"Sur le Gradient de la Politique pour les Systèmes Multi-Agents Coopératifs"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-01773969/","section":"Publications","summary":"","title":"Towards Executable Specifications for Microservices"},{"content":" Scientific lead: LEFORT Mathieu Is SyCoSMA the project lead? Yes Start date: 2017-10-04 End date: 2022-09-29 #Vous pouvez écrire une description ici\u0026hellip;\nLa fusion de données est une question cruciale en intelligence artificielle et en sciences cognitives. En effet, elle doit être résolue par tout agent, biologique ou artificiel, qui perçoit son environnement à travers plusieurs capteurs différenciés et complémentaires. Ce projet vise à mieux comprendre la perception chez l’humain afin d’améliorer les systèmes d’IA pour des agents plus autonomes. Il se positionne dans le cadre des théories sensori-motrices qui s’appuient sur un certain nombre d’indices provenant des neurosciences, de la psychologie, de la philosophie, \u0026hellip; Plus particulièrement, nous considérons la théorie des contingences sensori-motrices qui propose un cadre unifié fondé sur 2 principes : 1- L’apprentissage de corrélations sensori-motrices qui pose en particulier la question de l’intégration multi-sensorielle, dès lors que l’on considère plusieurs capteurs, et même multimodale en considérant l’action. Une des questions centrales est de savoir quand et comment fusionner les informations disponibles de manière autonome en s’appuyant par exemple sur l’apprentissage passé des relations entre les différentes données. 2- La perception active qui formalise que l’être humain interagit avec son environnement non seulement pour atteindre ses objectifs ou maintenir son intégrité physique (action pragmatique), mais également afin d’obtenir plus d’informations sur son environnement (action épistémique), par exemple via les saccades oculaires. Cela nécessite à la fois de disposer d’un modèle prédictif de la relation à l’environnement (en conjonction avec l’apprentissage de représentation sensori-motrices) et de décider dans quelle(s) modalité(s) et dans quelle(s) zone(s) de l’espace aller chercher la prochaine information pertinente.\nNous avons obtenu les résultats suivants :\nExpériences de psychophysique basées sur un paradigme de ventriloquie. On présente aux sujets un stimulus visuel (patch de Gabor de différents niveaux de contraste ou nuages de points avec densité variable) et un stimulus auditif (bruit rose de différents niveaux d’intensité dans un bruit blanc constant). Les sujets produisent des erreurs de localisation du son en direction du stimulus visuel (nommées biais crossmodal) si celui-ci est présenté de manière synchrone mais à une position spatiale différente du stimulus auditif. En particulier, nous avons mesuré sur 45 participants les effets de la fiabilité des stimuli sensoriels et des saccades oculaires (dans une perspective de perception active) sur le biais crossmodal. Les premières analyses statistiques de ces données. Elles montrent que l’augmentation de la fiabilité du stimulus visuel (resp. auditif) augmente (resp. diminue) le biais crossmodal et donc la capture visuelle. Les conclusions tirées des données expérimentales ont été affinées et renforcées par l’utilisation de modèles statistiques linéaires mixtes, ajustés sur des dizaines de milliers d’observations, permettant de mieux prendre en compte la variabilité des comportements humains, dont les différences inter-individuelles de sensibilité aux stimuli auditifs, visuels, ou d’usage des mouvements oculaires. La modélisation de la fusion audio-visuelle sur des données de la littérature. L’originalité réside dans l’utilisation du paradigme des champs neuronaux dynamiques (DNF) qui permettent à la fois de modéliser l’activité neuronale par champ moyen mais aussi de reproduire d’un point de vue comportemental des phénomènes saccadiques. Ils offrent ainsi des possibilités de modélisation plus riche et fine des mécanismes décisionnels, avec la capacité de simuler la dynamique de décisions individuelles, à l’inverse de modèles descriptifs s’intéressant davantage aux distributions sur des centaines d’essais. ","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/projects/18_amplifier/","section":"Axes","summary":"","title":"AMPLIFIER"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/tags/developmental-learning/","section":"Tags","summary":"","title":"Developmental Learning"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/tags/empathy/","section":"Tags","summary":"","title":"Empathy"},{"content":" Scientific lead: HASSAS Salima Is SyCoSMA the project lead? Yes Start date: 2017-01-07 End date: 2020-01-07 Cette thèse vise à concevoir et à développer des mécanismes permettant aux robots d’interaction d’être plus empathiques, et ce dans le but d’améliorer les interactions Homme-robot. L’empathie a été l’une des clés de l’évolution humaine et des interactions sociales et certaines régions du cerveau humain sont spécialisées dans la compréhension des émotions, des buts, des connaissances et des croyances des autres. Ce besoin d’échanges entre l’Homme et les robots s’est récemment traduit par le développement de robots spécialisés dans les interactions. Néanmoins, les robots actuellement disponibles sur le marché offrent un spectre d’interaction limité. Ces limitations s’expliquent par plusieurs facteurs : environnement bruité, peu ou pas structuré, mauvaise compréhension orale, mauvaise perception, événements imprévus, etc. Dans cette thèse, nous projetons d’utiliser l’Intelligence Artificielle (IA) développementale pour construire le cœur d’un moteur d’interaction capable de pallier ces limites en construisant des interactions qui sembleront plus empathiques.\nL’IA développementale est un paradigme de l’IA qui s’inspire notamment des travaux en psychologie développementale de Jean Piaget et qui fait récemment l’objet d’un intérêt croissant dans la communauté scientifique en général, et en IA pour la robotique en particulier. En effet, puisqu’elle permet un apprentissage progressif de concepts à partir de régularités d’interaction avec l’environnement observé, l’IA développementale se montre souvent plus adaptée aux types de problèmes que l’on rencontre en robotique. Contrairement à certaines approches qui requièrent une modélisation a priori du domaine, ou bien un gros volume de données pour constituer une base d’apprentissage, l’IA développementale est capable d’obtenir des résultats satisfaisants dans des temps raisonnables avec peu de données et dans des environnements qui ne sont pas connus à l’avance. Dans cette thèse, nous nous concentrerons donc sur la création d’interactions, que l’on souhaite le plus empathiques possibles, en nous basant sur une stratégie inspirée de l’IA développementale.\n","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/projects/22_hoomano/","section":"Axes","summary":"","title":"HOOMANO"},{"content":" Scientific lead: HASSAS Salima Is SyCoSMA the project lead? Yes Start date: 2017-01-02 End date: 2020-01-31 Le Laboratoire Commun ​ BEHAVIORS.AI (​Behaviors.ai is an Engine enHancing verbAl and non­Verbal InteractiOns of RobotS, based on Artificial Intelligence) a pour ambition de concevoir et développer un moteur d’interactions intelligent visant à rendre les interactions homme­robot plus empathiques, intuitives et naturelles. Il rassemble les compétences et savoir­faire des chercheurs en Intelligence Artificielle du LIRIS et de la société Hoomano, spécialisée dans le développement de logiciels pour les robots d’interaction. BEHAVIORS.AI permettra d’explorer les opportunités de transfert des travaux de recherche fondamentale, en particulier en ​ apprentissage développemental​, vers des applications de robotique destinées au grand public. Les ​ objectifs du projet BEHAVIORS.AI sont d’améliorer la perception du contexte et la façon d’y répondre, afin de tendre vers un mode d’interaction plus empathique entre les humains et le robot. Ceci passe par la construction de mécanismes d’interprétation des interactions s’appuyant à la fois sur les éléments de communication verbale et non­verbale : gestes, attitudes, expressions faciales, émotions, regards, informations de contexte. Il convient donc de repenser l’expérience utilisateur (UX) dans un contexte robotique, avec toutes les conséquences méthodologiques que cela peut avoir (agilité de conception interactionnelle pour un meilleur couplage humain­machine). Le projet ne vise pas à améliorer l’une ou l’autre des techniques de perceptions, mais à proposer des stratégies intelligentes pour les combiner et réagir, conduisant ainsi à des interactions plus pertinentes et plus réalistes, témoignant d’une meilleure adaptabilité du robot à son environnement. Le ​ projet de recherche vise à élaborer un moteur d’interaction capable non seulement d’utiliser des travaux de l’état de l’art mais aussi de développer de nouvelles solutions. Un accent particulier est mis sur des recherches très prometteuses en ​ apprentissage développemental et en ​ robotique cognitive​ . Les travaux s’inscrivent dans la ​ démarche d’innovation d’Hoomano qui consiste à produire des outils génériques et utilisables sur n’importe quel type de plateforme robotique, de sorte à devenir un moteur standard sur le marché​ . Outre les ​ défis techniques liés à l’hétérogénéité des plateformes, cette démarche soulève un ​ défi de recherche majeur : le moteur doit être capable de s’adapter dynamiquement aux capacités d’interactions disponibles sur le robot, et de se se développer perpétuellement en fonction de ses interactions intéressantes et effectives avec son environnement.\n","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/projects/19_behavior/","section":"Axes","summary":"","title":"BEHAVIORS.AI"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/tags/humar-robot-interaction/","section":"Tags","summary":"","title":"Humar-Robot Interaction"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-01967864/","section":"Publications","summary":"","title":"3D-SSF: A Bio-inspired Approach for Dynamic Multi-Subject Clustering of White Matter Tracts"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-01583031/","section":"Publications","summary":"","title":"A constraint-based coordination model to advantage buses in urban traffic"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-01839427/","section":"Publications","summary":"","title":"Active Multisensory Perception and LearnIng For InteractivE Robots"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-01578976/","section":"Publications","summary":"","title":"Ajustement diffus et adaptatif de la consommation électrique résidentielle par un système multi-agent auto-adaptatif"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/amira-chekir/","section":"Authors","summary":"","title":"Amira Chekir"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-01588822/","section":"Publications","summary":"","title":"An Authoring Tool to Elicit Knowledge to be Taught without Programming"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/arame-ndeye-diago/","section":"Authors","summary":"","title":"Arame Ndeye Diago"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/aurelien-vialon/","section":"Authors","summary":"","title":"Aurelien Vialon"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-01632771/","section":"Publications","summary":"","title":"Autonomous object modeling and exploiting: a new approach based on affordances from continual interaction with environment"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-01630163/","section":"Publications","summary":"","title":"Autonomous object recognition in videos using Siamese Neural Networks"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/awa-diattara/","section":"Authors","summary":"","title":"Awa Diattara"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/baudouin-dafflon/","section":"Authors","summary":"","title":"Baudouin Dafflon"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/christine-solnon/","section":"Authors","summary":"","title":"Christine Solnon"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-01576351/","section":"Publications","summary":"","title":"Classification des problèmes stochastiques et dynamiques de collectes et de livraisons par des véhicules intelligents"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-01637286/","section":"Publications","summary":"","title":"Co-construction of adaptive public policies using SmartGov"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-02076396/","section":"Publications","summary":"","title":"Comparaison de différents modèles de programmation par contraintes pour le clustering conceptuel"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-01565620/","section":"Publications","summary":"","title":"Contrôle par apprentissage constructiviste pour le trafic coopératif"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-01588175/","section":"Publications","summary":"","title":"Distributed negotiation for collective decision-making"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/eleftherios-garyfallidis/","section":"Authors","summary":"","title":"Eleftherios Garyfallidis"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-01591669/","section":"Publications","summary":"","title":"Extracting tags from large raw texts using End-to-End memory networks"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/fatima-oulebsir-boumghar/","section":"Authors","summary":"","title":"Fatima Oulebsir Boumghar"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/feras-al-kassar/","section":"Authors","summary":"","title":"Feras Al Kassar"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/franck-gechter/","section":"Authors","summary":"","title":"Franck Gechter"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/joanna-jongwane/","section":"Authors","summary":"","title":"Joanna Jongwane"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/kenji-tei/","section":"Authors","summary":"","title":"Kenji Tei"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-01616343/","section":"Publications","summary":"","title":"L'intelligence artificielle pour apprendre aux robots"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-01574683/","section":"Publications","summary":"","title":"Little AI: Playing a constructivist robot"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/marc-alexandre-c%C3%B4t%C3%A9/","section":"Authors","summary":"","title":"Marc-Alexandre Côté"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/matthis-gaciarz/","section":"Authors","summary":"","title":"Matthis Gaciarz"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/maxime-chabert/","section":"Authors","summary":"","title":"Maxime Chabert"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/maxime-descoteaux/","section":"Authors","summary":"","title":"Maxime Descoteaux"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/tel-01982580/","section":"Publications","summary":"","title":"Mécanismes de négociation distribuée pour la gestion intelligente de l’énergie"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-01553064/","section":"Publications","summary":"","title":"Multi-robot human scene observation based on hybrid metric-topological mapping"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-01526089/","section":"Publications","summary":"","title":"Multi-Robot Navigation and Cooperative Mapping in a Circular Topology"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/nawel-medjkoune/","section":"Authors","summary":"","title":"Nawel Medjkoune"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/pierre-antoine-champin/","section":"Authors","summary":"","title":"Pierre-Antoine Champin"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/romain-cailliere/","section":"Authors","summary":"","title":"Romain Cailliere"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/saber-mansour/","section":"Authors","summary":"","title":"Saber Mansour"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/simon-bultmann/","section":"Authors","summary":"","title":"Simon Bultmann"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-01567275/","section":"Publications","summary":"","title":"SmartGov : architecture générique pour exploiter les données des smart cities et co-concevoir des politiques crédibles"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-01519502/","section":"Publications","summary":"","title":"Soft-Goal Approximation Context Awareness of Goal-driven Self-Adaptive Systems"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/st%C3%A9phane-dalu/","section":"Authors","summary":"","title":"Stéphane D'Alu"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-01677247/","section":"Publications","summary":"","title":"Using Physics Inspired Wave Agents in a Virtual Environment: Longitudinal Distance Control in Robots Platoon"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/vanda-luengo/","section":"Authors","summary":"","title":"Vanda Luengo"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/vassilissa-lehoux/","section":"Authors","summary":"","title":"Vassilissa Lehoux"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-01371774/","section":"Publications","summary":"","title":"A constructivist approach for a self-adaptive decision-making system: application to road traffic control"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-01233860/","section":"Publications","summary":"","title":"A Self-Acquiring Knowledge Process for MCTS"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-01579098/","section":"Publications","summary":"","title":"Ajustement Diffus et Adaptatif de la Consommation Électrique Résidentielle par un Système Multi-Agents"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-01338916/","section":"Publications","summary":"","title":"Anticipation Based on a Bi-Level Bi-Objective Modeling for the Decision-Making in the Car-Following Behavior"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/antoine-nongaillard/","section":"Authors","summary":"","title":"Antoine Nongaillard"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/audrey-serna/","section":"Authors","summary":"","title":"Audrey Serna"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-01371894/","section":"Publications","summary":"","title":"Autonomous affordance construction without planning for environment-agnostic agents"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/bertrand-t.-david/","section":"Authors","summary":"","title":"Bertrand T. David"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/bouchareb-nassima/","section":"Authors","summary":"","title":"Bouchareb Nassima"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/dani%C3%A8le-patier/","section":"Authors","summary":"","title":"Danièle Patier"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-01355874/","section":"Publications","summary":"","title":"Decentralized and fair multilateral negotiation"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/eric-duchene/","section":"Authors","summary":"","title":"Eric Duchene"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/fahem-kebair/","section":"Authors","summary":"","title":"Fahem Kebair"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-01489782/","section":"Publications","summary":"","title":"Flexible Load Shedding Using Gossip Communication in a Multi-agents System"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-01496315/","section":"Publications","summary":"","title":"Freight Distribution Based on Delivery Area Booking"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-01298573/","section":"Publications","summary":"","title":"How to assess the benefits of connected vehicles? A simulation framework for the design of cooperative traffic management strategies"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-01357857/","section":"Publications","summary":"","title":"Incremental and adaptive multi-robot mapping for human scene observation"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-01330520/","section":"Publications","summary":"","title":"Intelligent Mechanism for Cloud Federation and Requirements Changes Management."},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/jan-pressler/","section":"Authors","summary":"","title":"Jan Pressler"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/jean-baptiste-thebaud/","section":"Authors","summary":"","title":"Jean-Baptiste Thebaud"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/jonathan-cohen/","section":"Authors","summary":"","title":"Jonathan Cohen"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/j%C3%B6rn-hurtienne/","section":"Authors","summary":"","title":"Jörn Hurtienne"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/julien-monteil/","section":"Authors","summary":"","title":"Julien Monteil"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/lionel-m%C3%A9dini/","section":"Authors","summary":"","title":"Lionel Médini"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/tags/load-shedding/","section":"Tags","summary":"","title":"Load Shedding"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-01335659/","section":"Publications","summary":"","title":"LOLLIPOP: Generating and using proper plan and negative refinements for online partial order planning"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-01329606/","section":"Publications","summary":"","title":"Managing energy markets in future smart grids using bilateral contracts"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/maxime-gueriau/","section":"Authors","summary":"","title":"Maxime Gueriau"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-01314543/","section":"Publications","summary":"","title":"Modèle de formation de coalitions A-core dans un contexte incertain"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-01261596/","section":"Publications","summary":"","title":"Multi-agent mechanism for efficient cooperative use of energy"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/nacer-eddine-zarour/","section":"Authors","summary":"","title":"Nacer Eddine Zarour"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/nam-tung-ly/","section":"Authors","summary":"","title":"Nam Tung Ly"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/pascal-fran%C3%A7ois-mbissane-faye/","section":"Authors","summary":"","title":"Pascal François Mbissane Faye"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/pierre-descombes/","section":"Authors","summary":"","title":"Pierre Descombes"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/ramdane-maamri/","section":"Authors","summary":"","title":"Ramdane Maamri"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/tel-01488151/","section":"Publications","summary":"","title":"Régulation de trafic urbain multimodal : une modélisation multi-agents"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/remy-rojas/","section":"Authors","summary":"","title":"Remy Rojas"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/ren%C3%A9-chalon/","section":"Authors","summary":"","title":"René Chalon"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-01160356/","section":"Publications","summary":"","title":"Resource management policies to increase provider’s gain in a Cloud Coalition"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/robert-tscharn/","section":"Authors","summary":"","title":"Robert Tscharn"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/romain-cailli%C3%A8re/","section":"Authors","summary":"","title":"Romain Caillière"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-01341812/","section":"Publications","summary":"","title":"Smart Lighting in Dementia Care Facility"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/stephan-huber/","section":"Authors","summary":"","title":"Stephan Huber"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-01254503/","section":"Publications","summary":"","title":"Stochastic Petri Net-Based Modeling and Formal Analysis of Fault Tolerant Contract Net Protocol"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/tel-01504421/","section":"Publications","summary":"","title":"Systèmes multi-agents, auto-organisation et contrôle par apprentissage constructiviste pour la modélisation et la régulation dans les systèmes coopératifs de trafic"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-01515382/","section":"Publications","summary":"","title":"Toward Constrained Semantic WoT"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-01357997/","section":"Publications","summary":"","title":"Towards an Authoring Tool to Acquire Knowledge for ITS Teaching Problem Solving Methods"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-01338338/","section":"Publications","summary":"","title":"Towards Supporting Caregivers to Monitor the Whereabouts of People with Dementia"},{"content":" Scientific lead: HASSAS Salima Is SyCoSMA the project lead? Yes Start date: 2016-01-01 End date: 2019-01-10 The field of Energy Management Systems for Smart Grids has been extensively explored in recent years, with many different approaches being described in the literature. In collaboration with our industrial partner Ubiant, which deploys smart homes solutions, we identified a need for a highly robust and scalable system that would exploit the flexibility of residential consumption to optimize energy use in the smart grid. At the same time we observed that the majority of existing works focused on the management of production and storage only, and that none of the proposed architectures are fully decentralized. Our objective was then to design a dynamic and adaptive mechanism to leverage every existing flexibility while ensuring the user\u0026rsquo;s comfort and a fair distribution of the load balancing effort ; but also to offer a modular and open platform with which a large variety of devices, constraints and even algorithms could be interfaced. In this thesis we realised (1) an evaluation of state of the art techniques in real-time individual load forecasting, whose results led us to follow (2) a bottom-up and decentralized approach to distributed residential load shedding system relying on a dynamic compensation mechanism to provide a stable curtailment. On this basis, we then built (3) a generic user-centered platform for energy management in smart grids allowing the easy integration of multiple devices, the quick adaptation to changing environment and constraints, and an efficient deployment.\n","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/projects/7_ubiant/","section":"Axes","summary":"","title":"Ubiant"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-01575872/","section":"Publications","summary":"","title":"WAVE: Wave Agents in a Virtual Environment; a Proof of Concept Applied to Longitudinal Distance Regulation in Platoon Control"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-01160409/","section":"Publications","summary":"","title":"A continuous negotiation based model for traffic regulation at an intersection"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/abdel-illah-mouaddib/","section":"Authors","summary":"","title":"Abdel-Illah Mouaddib"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/tel-01213353/","section":"Publications","summary":"","title":"Approche décentralisée de l'apprentissage constructiviste et modélisation multi-agent du problème d'amorçage de l'apprentissage sensorimoteur en environnement continu : application à l'intelligence ambiante"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-01146381/","section":"Publications","summary":"","title":"Approche décentralisée pour un apprentissage constructiviste en environnement continu : application à l'intelligence ambiante."},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-01184566/","section":"Publications","summary":"","title":"ARTIFICIAL FINANCIAL MARKET. Risk Analysis Approach"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-01134237/","section":"Publications","summary":"","title":"Automated Negotiation for Traffic Regulation"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-01246256/","section":"Publications","summary":"","title":"Autonomous object modeling based on affordances in a dynamic environment"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/badiaa-hedjazi/","section":"Authors","summary":"","title":"Badiaa Hedjazi"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-01254964/","section":"Publications","summary":"","title":"Concentric and Incremental Multi-Robot Mapping to Observe Complex Scenes"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-01184565/","section":"Publications","summary":"","title":"Constraint-based negotiation model for traffic regulation"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-01162679/","section":"Publications","summary":"","title":"Continuous Negotiation for a Vehicle-Regulated Intersection"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-01200755/","section":"Publications","summary":"","title":"Decentralized multi-robot planning to explore and perceive"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-01184564/","section":"Publications","summary":"","title":"Dynamic Coalitions formation in Dynamic Uncertain Environments"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/tel-01234642/","section":"Publications","summary":"","title":"Dynamique d'apprentissage pour Monte Carlo Tree Search : applications aux jeux de Go et du Clobber solitaire impartial"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/karima-benatchba/","section":"Authors","summary":"","title":"Karima Benatchba"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/laurent-jeanpierre/","section":"Authors","summary":"","title":"Laurent Jeanpierre"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/tel-01167110/","section":"Publications","summary":"","title":"Modèles de formation de coalitions stables dans un contexte ad-hoc et stochastique"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-01251602/","section":"Publications","summary":"","title":"Modeling Biological Agents Beyond the Reinforcement Learning Paradigm"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-01214296/","section":"Publications","summary":"","title":"Multi-Agent Dynamic Coupling for Cooperative Vehicles Modeling"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/neila-bhouri/","section":"Authors","summary":"","title":"Neila Bhouri"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-01197468/","section":"Publications","summary":"","title":"Nested Monte-Carlo Search for Multi-Agent Coalitions Mechanism With Constraints"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/tel-01236089/","section":"Publications","summary":"","title":"On enhancing recommender systems by utilizing general social networks combined with users goals and contextual awareness"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/pascal-fran%C3%A7ois-faye/","section":"Authors","summary":"","title":"Pascal François Faye"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/rana-chamsi-abu-quba/","section":"Authors","summary":"","title":"Rana Chamsi Abu Quba"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/r%C3%A9mi-casado/","section":"Authors","summary":"","title":"Rémi Casado"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/s%C3%A9bastien-mazac/","section":"Authors","summary":"","title":"Sébastien Mazac"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-01193348/","section":"Publications","summary":"","title":"Towards Intelligent and Implicit Assistance for People with Dementia: Support for Orientation and Navigation"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/tristan-cazenave/","section":"Authors","summary":"","title":"Tristan Cazenave"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-01163018/","section":"Publications","summary":"","title":"Un simulateur multiagent de trafic coopératif"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/halshs-01279230/","section":"Publications","summary":"","title":"A multiagent multilateral negotiation protocol for joint decision-making"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-01267713/","section":"Publications","summary":"","title":"A self-organized system improving inner topology for data sharing efficiency"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-01270904/","section":"Publications","summary":"","title":"Adapting Agent's Interactions in Dynamic Contexts."},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-01313180/","section":"Publications","summary":"","title":"An extension of MovSim for Multi-Agent Cooperative Vehicles Modeling"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-01301061/","section":"Publications","summary":"","title":"Bi-level token ring for multilateral negotiations"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/chantal-berdier/","section":"Authors","summary":"","title":"Chantal Berdier"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/christine-gertosio/","section":"Authors","summary":"","title":"Christine Gertosio"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-01497823/","section":"Publications","summary":"","title":"Decentralized Multi-Robot Planning to Explore and Perceive"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-01301097/","section":"Publications","summary":"","title":"Dynamiques, Couplages et Visions Intégratives des Systèmes Multi-Agents, Numéro spécial de la Revue d'Intelligence Artificielle - RSTI série RIA- Volume 28 - N° 4/2014"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/fayyad-usama/","section":"Authors","summary":"","title":"Fayyad Usama"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-01598607/","section":"Publications","summary":"","title":"Formation de coalition stable dans un contexte non-déterministe et instable"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-01301057/","section":"Publications","summary":"","title":"Freight distribution based on delivery area booking"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-01301058/","section":"Publications","summary":"","title":"From a “Cold” to a “Warm” Start in Recommender systems"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/hamamache-kheddouci/","section":"Authors","summary":"","title":"Hamamache Kheddouci"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/hammam-chamsi/","section":"Authors","summary":"","title":"Hammam Chamsi"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-01552137/","section":"Publications","summary":"","title":"iSoNTRE : The Intelligent Social Network Transformer into Recommendation Engine framework"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-01552136/","section":"Publications","summary":"","title":"iSoNTRE: the Social Network Transformer into Recommendation Engine"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-01083449/","section":"Publications","summary":"","title":"Knowledge complement for Monte Carlo Tree Search: an application to combinatorial games"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-01329655/","section":"Publications","summary":"","title":"Modeling a Multi-issue Negotiation Protocol for Agent Extensible Negotiations"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/mohammed-haddad/","section":"Authors","summary":"","title":"Mohammed Haddad"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-01135987/","section":"Publications","summary":"","title":"On bootstrapping sensori-motor patterns for a constructivist learning system in continuous environments"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/s.-arib/","section":"Authors","summary":"","title":"S. Arib"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-01329648/","section":"Publications","summary":"","title":"Stabilizing Agent's Interactions in Dynamic Contexts"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/publications/hal-01598606/","section":"Publications","summary":"","title":"Stable collaboration in unstable environments"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/authors/usama-fayyad/","section":"Authors","summary":"","title":"Usama Fayyad"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/members/2674/","section":"Members","summary":"","title":"Alexandre Devillers"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/members/3154/","section":"Members","summary":"","title":"Alice Janela Cameijo"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/statuses/associate-professor-/","section":"Statuses","summary":"","title":"Associate Professor "},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/employers/autre/","section":"Employers","summary":"","title":"Autre"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/members/3185/","section":"Members","summary":"","title":"Axel Bessy"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/members/3045/","section":"Members","summary":"","title":"Bruno Yun"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/employers/centre-national-de-la-recherche-scientifique/","section":"Employers","summary":"","title":"Centre National De La Recherche Scientifique"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/statuses/collaborateur-ext%C3%A9rieur-/","section":"Statuses","summary":"","title":"Collaborateur Extérieur "},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/employers/%C3%A9cole-centrale-de-lyon/","section":"Employers","summary":"","title":"École Centrale De Lyon"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/members/3846/","section":"Members","summary":"","title":"Emilien Komlenovic"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/employers/","section":"Employers","summary":"","title":"Employers"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/members/3325/","section":"Members","summary":"","title":"Farah Ben Slama"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/members/281/","section":"Members","summary":"","title":"Frédéric Armetta"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/locations/hors-liris/","section":"Locations","summary":"","title":"Hors LIRIS"},{"content":"Locations # Nautibus (Université Lyon 1) Jobs # Internship opportunities at LIRIS: https://liris.cnrs.fr/emplois/offres-de-stage ","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/info/","section":"","summary":"","title":"Info"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/employers/institut-national-des-sciences-appliqu%C3%A9es-de-lyon/","section":"Employers","summary":"","title":"Institut National Des Sciences Appliquées De Lyon"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/members/3397/","section":"Members","summary":"","title":"Julien Lefebvre"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/members/367/","section":"Members","summary":"","title":"Laetitia Matignon"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/members/3343/","section":"Members","summary":"","title":"Laury Diadhiou--Coulon"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/locations/","section":"Locations","summary":"","title":"Locations"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/members/3579/","section":"Members","summary":"","title":"Louis Bagot"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/members/471/","section":"Members","summary":"","title":"Mathieu Lefort"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/members/3342/","section":"Members","summary":"","title":"Maxime Morge"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/members/2925/","section":"Members","summary":"","title":"Maya Medjad"},{"content":" ","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/members/","section":"Members","summary":"","title":"Members"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/members/3429/","section":"Members","summary":"","title":"Mohamed Massamba Sene"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/members/3170/","section":"Members","summary":"","title":"Nathan Salazar"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/locations/nautibus-universit%C3%A9-lyon1/","section":"Locations","summary":"","title":"Nautibus (Université Lyon1)"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/members/2253/","section":"Members","summary":"","title":"Nicolas Jacquelin"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/members/291/","section":"Members","summary":"","title":"Olivier Georgeon"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/statuses/phd-student-/","section":"Statuses","summary":"","title":"PhD Student "},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/statuses/post-doctoral-tenure-/","section":"Statuses","summary":"","title":"Post-Doctoral Tenure "},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/statuses/professor-/","section":"Statuses","summary":"","title":"Professor "},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/members/2199/","section":"Members","summary":"","title":"Rémy Chaput"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/locations/saint-%C3%A9tienne-ecl/","section":"Locations","summary":"","title":"Saint-Étienne (ECL)"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/members/359/","section":"Members","summary":"","title":"Samir Aknine"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/statuses/","section":"Statuses","summary":"","title":"Statuses"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/statuses/teaching-assistant-/","section":"Statuses","summary":"","title":"Teaching Assistant "},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/members/3760/","section":"Members","summary":"","title":"Télio Dupuis"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/themes/","section":"Themes","summary":"","title":"Themes"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/members/3072/","section":"Members","summary":"","title":"Timon Deschamps"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/statuses/trainee-/","section":"Statuses","summary":"","title":"Trainee "},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/employers/universit%C3%A9-claude-bernard-lyon-1/","section":"Employers","summary":"","title":"Université Claude Bernard Lyon 1"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/members/3424/","section":"Members","summary":"","title":"Valentin Cuzin-Rambaud"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/members/3818/","section":"Members","summary":"","title":"Valentin Laporte"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/members/362/","section":"Members","summary":"","title":"Véronique Deslandres"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/members/3851/","section":"Members","summary":"","title":"Weikang Zeng"},{"content":"","date":null,"permalink":"https://sycosma.projet.liris.cnrs.fr/members/3352/","section":"Members","summary":"","title":"Yasmine Bouamra"}]