ELSIM: End-to-end learning of reusable skills through intrinsic motivation

Autor: Salima Hassas, Laëtitia Matignon, Arthur Aubret

Publikováno v: Machine Learning and Knowledge Discovery in Databases ISBN: 9783030676605
ECML/PKDD (2)
Journées Francophones sur la Planification, la Décision et l'Apprentissage pour la conduite de systèmes
Journées Francophones sur la Planification, la Décision et l'Apprentissage pour la conduite de systèmes, Jul 2020, Angers, France
European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases (ECML-PKDD)
European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases (ECML-PKDD), Sep 2020, Ghent, Belgium
4th Lifelong Learning Workshop at ICML 2020
4th Lifelong Learning Workshop at ICML 2020, Jul 2020, Vienne, Austria

Taking inspiration from developmental learning, we present a novel reinforcement learning architecture which hierarchically learns and represents self-generated skills in an end-to-end way. With this architecture, an agent focuses only on task-reward

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_dedup___::a0143a5ed116ddb244119fc104fa1264

Context Aware Robot Architecture, Application to the RoboCup@Home Challenge

Autor: Raphael Leber, Laëtitia Matignon, Christian Wolf, Fabrice Jumel, Jacques Saraydaryan, Eric Lombardi, Olivier Simonin

Publikováno v: RoboCup 2018: Robot World Cup XXII ISBN: 9783030275433
RoboCup
LNAI Lecture Notes in Artificial Intelligence, proceedings of Robocup symposium 2018
RoboCup symposium
RoboCup symposium, Jun 2018, Montreal, Canada. pp.1-12, ⟨10.1007/978-3-030-27544-0_17⟩
RoboCup symposium, Jun 2018, Montreal, Canada. pp.1-12

International audience; This paper presents an architecture dedicated to the orchestration of high level abilities of a humanoid robot, such as a Pepper, which must perform some tasks as the ones proposed in the RoboCup@Home competition. We present t

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_dedup___::5add548271ae12d60fbf0831eb3ee2fd
https://doi.org/10.1007/978-3-030-27544-0_17

Cooperative Multi-Agent Policy Gradient

Autor: Guillaume Bono, Florian Pereyron, Jilles Steeve Dibangoye, Olivier Simonin, Laëtitia Matignon

Publikováno v: ECML-PKDD 2018-European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases
ECML-PKDD 2018-European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases, Sep 2018, Dublin, Ireland. pp.459-476
Machine Learning and Knowledge Discovery in Databases ISBN: 9783030109240
ECML/PKDD (1)
ECML-PKDD 2018-European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases, Sep 2018, Dublin, Ireland. pp.459-476, ⟨10.1007/978-3-030-10925-7_28⟩

Reinforcement Learning (RL) for decentralized partially observable Markov decision processes (Dec-POMDPs) is lagging behind the spectacular breakthroughs of single-agent RL. That is because assumptions that hold in single-agent settings are often obs

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_dedup___::95f8a4abd58e134d881f7606bfa39378
https://hal.inria.fr/hal-01847775

Independent reinforcement learners in cooperative Markov games: a survey regarding coordination problems

Autor: Laëtitia Matignon, Guillaume J. Laurent, Nadine Le Fort-Piat

Publikováno v: Knowledge Engineering Review
Knowledge Engineering Review, Cambridge University Press (CUP), 2012, 27 (1), pp.1-31. ⟨10.1017/S026988891200057⟩

In the framework of fully cooperative multi-agent systems, independent (non-communicative) agents that learn by reinforcement must overcome several difficulties to manage to coordinate. This paper identifies several challenges responsible for the non

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_dedup___::f5e6e63f5a9c13294321b6bf2e13c5e6
https://doi.org/10.1017/s0269888912000057

Decentralized multi-robot planning to explore and perceive

Autor: Abdel-Illah Mouaddib, Laëtitia Matignon, Laurent Jeanpierre

Publikováno v: Acta Polytechnica
Acta Polytechnica, Czech Technical University in Prague, 2015, 55 (3), pp.169-176. ⟨10.14311/AP.2015.55.0169⟩
HAL
Workshop on Multi-Agent Coordination in Robotic Exploration, ECAI 2014 (European Conference on Artificial Intelligence)
Workshop on Multi-Agent Coordination in Robotic Exploration, ECAI 2014 (European Conference on Artificial Intelligence), Aug 2014, Prague, Czech Republic
Acta Polytechnica, Vol 55, Iss 3, Pp 169-176 (2015)

International audience; In a recent French robotic contest, the objective was to develop a multi-robot system able to autonomously map and explore an unknown area while also detecting and localizing objects in the environment. The scientific issues o

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_dedup___::3cd71bfd750009211dd7516148d7f320
https://hal.archives-ouvertes.fr/hal-01200755