A voir également
- Short Bio
- Publications -----------------------------
- Researches
- News
- Code
- CV
- Teaching
- Cours Robotique Cogmaster
- Cours d'Esprit Critique
Teaching
Title : Research Director
Address : 4 place Jussieu, CC 173, 75252 Paris cedex 05
Phone : +33 (0) 1 44 27 28 85
Email : khamassi(at)isir.upmc.fr
Group : AMAC (AMAC)
Teaching
Example of a course given at the Brain, Cognition, Behavior and Technology summerschool, organized by Paul Verschure in Barcelona, in September 2019:
Teaching (current)
2015- |
Ecole Normale Supérieure Ulm, Paris, CogMaster Course coordinators: Mehdi Khamassi and Benoît Girard |
"Robotic modelling approaches to Cognitive Sciences" (click on the link to access to course material) |
2015- |
Université Paris Saclay (Orsay), Master 2 Information, Learning, Cognition Course coordinator: Michèle Sebag |
"Reinforcement learning models and their applications to Robotics and Neuroscience" |
2014- |
Sorbonne Université (ex UPMC), Licence (all disciplines) Course coordinators: Frédéric Decremps, Emmanuel Guigon and Mehdi Khamassi |
http://www.isir.upmc.fr/?op=view_profil&lang=fr&id=82&pageid=1300 |
2012- |
Ecole Normale Supérieure Ulm, Paris, Master 1 of Neuroscience Course Coordinators: Clément Léna and Daniela Popa |
"Reinforcement learning: computational models and model-based analysis of electrophysiological data" |
2011- |
Sorbonne Université (ex UPMC), International Master 2 of Mechatronics Systems for Rehabilitation (Robotics) Course coordinator: Véronique Perdereau |
"Bio-inspired action selection and learning" |
2010- |
Université Claude Bernard, Lyon 1, Master 2 of Integrative Neuroscience Course coordinators: Céline Amiez and Jean-Claude Dreher |
"Decision-making and reinforcement learning: elements of computational modelling" |
2008- |
Sorbonne Université (ex UPMC), Master 2 of Artificial Intelligence and Decision (Computer Science) then Master 2 ANDROIDE Course coordinators: Stéphane Doncieux then Nicolas Bredèche |
"Reinforcement learning: From neural processes modelling to robotics applications" |
Teaching (past)
2019 |
University of Allahabad (Prayagraj, India), GIAN Course Coordinateurs du cours : Narayanan Srinivasan, Chandrasekhar V.S. Parmi (RIP) et Mehdi Khamassi |
Advances in learning and decision-making |
2017 |
Univ. Pierre & Marie Curie (UPMC) – Paris 6, Master 2 Integrative Neuroscience Coordinateur du cours : Ouriel Gyrnszpan |
"Modèles computationnels de l'apprentissage et la prise de décision en contextes social et non-social |
2017 |
SMART Summerschool in Computational Social & Behav Sciences (UPMC) Coordinateur du cours : Mohamed Chetouani |
"Modèles computationnels de l'apprentissage par renforcement model-free et model-based |
2013-2016 |
Ecole Polytechnique, Palaiseau, Brain and Cognition Seminar Series Course coordinators: Yves Frégnac and Cyril Monier |
"Reinforcement learning models and their applications to Robotics and Neuroscience" |
2012-2015 |
Course coordinator: Jean-Baptiste Mouret |
"Reinforcement learning and neuro-inspired decision-making" |
2014 |
Ecole Normale Supérieure Ulm / PSL Labex, Paris, Master 2 d'Cognitive Engineering Course coordinators: Emmanuel Dupoux and Srdjan Ostojic |
"Reinforcement learning models and their applications to Robotics and Neuroscience" |
2013-2014 |
Université Claude Bernard, Lyon 1, Licence (all disciplines) Course coordinators: Denis Machon and Maïlys Faraut |
"Education to images and critical thinking in front of advertising" |
2013 |
Harvard Summer Class in Trento, Italy Course coordinator: Giorgio Coricelli |
"Reinforcement learning: computational models and model-based analysis of electrophysiological data" |
2013 |
Univ. Pierre & Marie Curie – Paris 6, Master 2 Integrative Neuroscience Course coordinator: Laure Rondi-Reig |
"Computational models of parallel memory systems for navigation" |
2012 |
Telluride Neuromorphic Engineering Summerschool, USA Summerschool directors: Ralph Etienne-Cummings, Timothy Horiuchi and Tobi Delbruck |
"Reinforcement learning models and their applications to Robotics and Neuroscience" |
Downloadable material
Sorbonne University 2019, Course on Reinforcement Learning models, their links with Neurobiology, and their applications to Autonomous Robotics:
-
Model-free (MF) Reinforcement Learning (RL)
- Temporal-Difference Learning
- Links with Neurobiological and Psychological data
- Applications of RL to the Neuroscience of decision-making
- Continuous state-space Reinforcement Learning in Robotics
- Limitations of MF RL in Neuroscience and Robotics
-
Model-based (MB) Reinforcement Learning (RL)
- Formalism (e.g. dyna-Q, dynamic programming)
- Off-line processing and the hippocampus
- Meta-Learning for the coordination of MF and MB RL
- Meta-Learning for the dynamic regulation of RL parameters
- Applications to Robotics and feedback to Neuroscience