Reinforcement learning

This course is an introduction to a branch of machine learning called reinforcement learning (RL). In this course, we'll look at the main models used in RL: multi-armed bandits, Markov decision processes, and their multi-agent and partial observation extensions, both in the dynamic framework and in the function approximation framework (by neural networks in particular). We will study the most important algorithms: value iteration, strategy iteration, Q-learning, DQN (Deep Q-learning). They will be implemented in Python.

This course is an introduction to a branch of machine learning called reinforcement learning (RL). In this course, we'll look at the main models used in RL: multi-armed bandits, Markov decision processes, and their multi-agent and partial observation extensions, both in the dynamic framework and in the function approximation framework (by neural networks in particular). We will study the most important algorithms: value iteration, strategy iteration, Q-learning, DQN (Deep Q-learning). They will be implemented in Python.