[{"command":"insert","method":null,"selector":".course-modal-ajax-placeholder-968997","data":"\u003Cdiv class=\u0022container content-rte\u0022\u003E\n  \u003Cdiv class=\u0022row\u0022\u003E\n    \u003Cdiv class=\u0022box col-md-6 bg-primary text-white\u0022\u003E\n      \u003Ch2 class=\u0022h1 text-left\u0022\u003E  Reinforcement Learning\n\u003C\/h2\u003E\n                  \u003Cdiv class=\u0022field-spacing\u0022\u003E\n        \u003Cdiv class=\u0022field__label h3 dash\u0022\u003EECTS :\n          2\n        \u003C\/div\u003E\n      \u003C\/div\u003E\n                  \u003Cdiv class=\u0022field-spacing\u0022\u003E\n        \u003Cdiv class=\u0022field__label h3 dash\u0022\u003ESemester:\n          Semestre 1\n        \u003C\/div\u003E\n      \u003C\/div\u003E\n            \u003Cdiv class=\u0022field-spacing\u0022\u003E\n      \u003Cdiv class=\u0022field__label h3 dash\u0022\u003ED\u00e9tail du volume horaire\u003C\/div\u003E\n                        \u003Cdiv class=\u0022mt-5 text-justify\u0022\u003E\n                          \u003Cspan class=\u0022field__label h4 mt-5 \u0022\u003ELecture:\u003C\/span\u003E\n                                      \u003Cspan\u003E9\u003C\/span\u003E\n                      \u003C\/div\u003E\n                                              \u003Cdiv class=\u0022mt-5 text-justify\u0022\u003E\n                          \u003Cspan class=\u0022field__label h4 mt-5 \u0022\u003EDirected study:\u003C\/span\u003E\n                                      \u003Cspan\u003E9\u003C\/span\u003E\n                      \u003C\/div\u003E\n                                                            \u003C\/div\u003E\n        \u003Cdiv class=\u0022field-spacing\u0022\u003E\n      \u003Cdiv class=\u0022field__label h3 dash\u0022\u003ELangue d\u0026#039;enseignement\u003C\/div\u003E\n                        \u003Cdiv class=\u0022mt-5 text-justify\u0022\u003E\n                                      \u003Cspan\u003EAnglais\u003C\/span\u003E\n                      \u003C\/div\u003E\n                  \u003C\/div\u003E\n        \u003Cdiv class=\u0022field-spacing\u0022\u003E\n      \u003Cdiv class=\u0022field__label h3 dash\u0022\u003EEnseignement \u00e0 distance\u003C\/div\u003E\n                        \u003Cdiv class=\u0022mt-5 text-justify\u0022\u003E\n                                      \u003Cspan\u003Enon\u003C\/span\u003E\n                      \u003C\/div\u003E\n                  \u003C\/div\u003E\n      \u003C\/div\u003E\n    \u003Cdiv class=\u0022box col-md-6\u0022\u003E\n            \u003Cdiv class=\u0022field-spacing\u0022\u003E\n      \u003Cdiv class=\u0022field__label h3 dash\u0022\u003EPr\u00e9requis\u003C\/div\u003E\n                        \u003Cdiv class=\u0022mt-5 text-justify\u0022\u003E\n                                      \u003Cspan\u003E\u003Cul\u003E\n\u003Cli\u003EKnowledge of Python\u003C\/li\u003E\n\u003Cli\u003EUnderstanding of probability\u0026nbsp;\u003C\/li\u003E\n\u003Cli\u003EBasic notions of deep learning\u003C\/li\u003E\n\u003C\/ul\u003E\u003C\/span\u003E\n                      \u003C\/div\u003E\n                  \u003C\/div\u003E\n        \u003Cdiv class=\u0022field-spacing\u0022\u003E\n      \u003Cdiv class=\u0022field__label h3 dash\u0022\u003EProgramme \/ plan \/ contenus\u003C\/div\u003E\n                        \u003Cdiv class=\u0022mt-5 text-justify\u0022\u003E\n                                      \u003Cspan\u003E\u003Cp\u003E\u003Cstrong\u003ESession 1: Introduction to Reinforcement Learning\u003C\/strong\u003E\u003C\/p\u003E\n\u003Cul\u003E\n\u003Cli\u003EWhat is RL?\u0026nbsp;\n\u003Cul\u003E\n\u003Cli\u003EDifferences between RL, supervised learning, and unsupervised learning\u0026nbsp;\u003C\/li\u003E\n\u003Cli\u003EApplications and use cases\u003C\/li\u003E\n\u003C\/ul\u003E\n\u003C\/li\u003E\n\u003Cli\u003EBasic terminology and notation\u003C\/li\u003E\n\u003Cli\u003EMarkov Decision Processes (MDPs)\n\u003Cul\u003E\n\u003Cli\u003EFormal framework of RL\u0026nbsp;\u003C\/li\u003E\n\u003Cli\u003EStates, actions, rewards, and transition dynamics\u0026nbsp;\u003C\/li\u003E\n\u003Cli\u003EPolicy and value functions\u003C\/li\u003E\n\u003C\/ul\u003E\n\u003C\/li\u003E\n\u003Cli\u003EDynamic programming concepts\u0026nbsp;\u003C\/li\u003E\n\u003Cli\u003EPolicy evaluation and improvement\u0026nbsp;\u003C\/li\u003E\n\u003Cli\u003EPolicy iteration\u0026nbsp;\u003C\/li\u003E\n\u003Cli\u003EValue iteration\u003C\/li\u003E\n\u003C\/ul\u003E\n\u003Cp\u003E\u003Cstrong\u003ESession 2: Tabular Methods\u003C\/strong\u003E\u003C\/p\u003E\n\u003Cul\u003E\n\u003Cli\u003EMonte Carlo methods\u0026nbsp;\u003C\/li\u003E\n\u003Cli\u003EMonte Carlo prediction and control\u0026nbsp;\u003C\/li\u003E\n\u003Cli\u003EExploration and initial visit\u0026nbsp;\u003C\/li\u003E\n\u003Cli\u003EEpisodic tasks\n\u003Cul\u003E\n\u003Cli\u003ETemporal difference (TD) learning\u0026nbsp;\u003C\/li\u003E\n\u003Cli\u003ESARSA and Q-learning\u003C\/li\u003E\n\u003C\/ul\u003E\n\u003C\/li\u003E\n\u003Cli\u003EOn-policy vs. off-policy learning\u0026nbsp;\u003C\/li\u003E\n\u003Cli\u003EBootstrapping\u003C\/li\u003E\n\u003C\/ul\u003E\n\u003Cp\u003E\u003Cstrong\u003ESession 3: Function approximation\u003C\/strong\u003E\u003C\/p\u003E\n\u003Cul\u003E\n\u003Cli\u003ELinear approximation\u0026nbsp;\u003C\/li\u003E\n\u003Cli\u003EGradient descent methods\u0026nbsp;\u003C\/li\u003E\n\u003Cli\u003EPolicy gradient methods\u0026nbsp;\n\u003Cul\u003E\n\u003Cli\u003EREINFORCE algorithm\u0026nbsp;\u003C\/li\u003E\n\u003Cli\u003EVariance reduction techniques\u003C\/li\u003E\n\u003C\/ul\u003E\n\u003C\/li\u003E\n\u003C\/ul\u003E\n\u003Cp\u003E\u003Cstrong\u003ESession 4: Deep Learning for the Value Function\u003C\/strong\u003E\u003C\/p\u003E\n\u003Cul\u003E\n\u003Cli\u003EQ-Networks (DQN) and developments\u003C\/li\u003E\n\u003Cli\u003EUse of replay buffers\u003C\/li\u003E\n\u003Cli\u003EDouble DQN\u003C\/li\u003E\n\u003C\/ul\u003E\n\u003Cp\u003E\u003Cstrong\u003ESession 5: Deep Learning for Policy\u003C\/strong\u003E\u003C\/p\u003E\n\u003Cul\u003E\n\u003Cli\u003EBasic concepts of actor-critic algorithms\u003C\/li\u003E\n\u003Cli\u003EProximal Policy Optimization (PPO)\u003C\/li\u003E\n\u003C\/ul\u003E\n\u003Cp\u003E\u003Cstrong\u003ESession 6: Advanced Topics and Recent Trends\u003C\/strong\u003E\u003C\/p\u003E\n\u003Cul\u003E\n\u003Cli\u003EMulti-agent RL\u0026nbsp;\u003C\/li\u003E\n\u003Cli\u003EExploration strategies\u003C\/li\u003E\n\u003Cli\u003EMCTS usage\u003C\/li\u003E\n\u003C\/ul\u003E\u003C\/span\u003E\n                      \u003C\/div\u003E\n                  \u003C\/div\u003E\n        \u003Cdiv class=\u0022field-spacing\u0022\u003E\n      \u003Cdiv class=\u0022field__label h3 dash\u0022\u003EObjectifs d\u0026#039;apprentissage\u003C\/div\u003E\n                        \u003Cdiv class=\u0022mt-5 text-justify\u0022\u003E\n                                      \u003Cspan\u003E\u003Cp\u003EThis course provides an in-depth introduction to reinforcement learning (RL), emphasizing fundamental algorithms and practical applications. Students will learn the principles of sequential decision-making, Markov decision processes, value functions, policy optimization, and some modern techniques in deep RL. Throughout the course, students will experiment the studied algorithms and approaches by implementing Python agents that learn to solve academic problems.\u003C\/p\u003E\n\u003Cp\u003EIn summary, the learning objectives are as follows:\u003C\/p\u003E\n\u003Cul\u003E\n\u003Cli\u003EUnderstand the fundamental concepts of reinforcement learning;\u003C\/li\u003E\n\u003Cli\u003EImplement classic Monte Carlo and temporal-difference (TD) algorithms, extending to actor-critic algorithms;\u003C\/li\u003E\n\u003Cli\u003EExplore function approximation and deep learning through RL.\u003C\/li\u003E\n\u003C\/ul\u003E\u003C\/span\u003E\n                      \u003C\/div\u003E\n                  \u003C\/div\u003E\n        \u003Cdiv class=\u0022field-spacing\u0022\u003E\n      \u003Cdiv class=\u0022field__label h3 dash\u0022\u003EOrganisation g\u00e9n\u00e9rale et modalit\u00e9s p\u00e9dagogiques\u003C\/div\u003E\n                        \u003Cdiv class=\u0022mt-5 text-justify\u0022\u003E\n                                      \u003Cspan\u003E\u003Cp\u003EThe course is organized into six three-hour sessions. During each session, students will apply the concepts they have learned by programming learning agents.\u003C\/p\u003E\u003C\/span\u003E\n                      \u003C\/div\u003E\n                  \u003C\/div\u003E\n        \u003Cdiv class=\u0022field-spacing\u0022\u003E\n      \u003Cdiv class=\u0022field__label h3 dash\u0022\u003EModalit\u00e9s p\u00e9dagogiques particuli\u00e8res\u003C\/div\u003E\n                        \u003Cdiv class=\u0022mt-5 text-justify\u0022\u003E\n                                      \u003Cspan\u003E\u003Cp\u003EThis course will be evaluated through a written theory exam and a project report.\u003C\/p\u003E\u003C\/span\u003E\n                      \u003C\/div\u003E\n                  \u003C\/div\u003E\n        \u003Cdiv class=\u0022field-spacing\u0022\u003E\n      \u003Cdiv class=\u0022field__label h3 dash\u0022\u003EComp\u00e9tences\u003C\/div\u003E\n                        \u003Cdiv class=\u0022mt-5 text-justify\u0022\u003E\n                                      \u003Cspan\u003E\u003Cp\u003EStudents will demonstrate their theoretical and applied knowledge of modern RL methods. The applied aspects will allow students to demonstrate their ability to program agents in Python using the PyTorch and StableBaseline libraries.\u003C\/p\u003E\u003C\/span\u003E\n                      \u003C\/div\u003E\n                  \u003C\/div\u003E\n        \u003Cdiv class=\u0022field-spacing\u0022\u003E\n      \u003Cdiv class=\u0022field__label h3 dash\u0022\u003EBibliographie\u003C\/div\u003E\n                        \u003Cdiv class=\u0022mt-5 text-justify\u0022\u003E\n                                      \u003Cspan\u003E\u003Cp\u003E\u0022Reinforcement Learning: An Introduction\u0022 par Sutton \u0026amp; Barto\u003C\/p\u003E\u003C\/span\u003E\n                      \u003C\/div\u003E\n                  \u003C\/div\u003E\n        \u003Cdiv class=\u0022field-spacing\u0022\u003E\n      \u003Cdiv class=\u0022field__label h3 dash\u0022\u003ENature de l\u0026#039;\u00e9valuation\u003C\/div\u003E\n                        \u003Cdiv class=\u0022mt-5 text-justify\u0022\u003E\n                                      \u003Cspan\u003EEvaluation Continue non Int\u00e9grale\u003C\/span\u003E\n                      \u003C\/div\u003E\n                  \u003C\/div\u003E\n      \u003C\/div\u003E\n  \u003C\/div\u003E\n\u003C\/div\u003E\n","settings":null}]