Webpage of the RL class @Centrale Lille (SDI)

This 20 hours Reinforcement Learning class is jointly taught with Omar Darwiche Domingues (Inria Scool). The slides of the course and notebooks for the practical sessions will be made online before the class.

The course will be illustated by practical sessions in Python, using jupyter notebook. To validate the class, you need to do a project (report due mid January, oral presentation end of January).

• Lecture 1: Markov Decision Processes. Illustrative notebooks.
• Lecture 2: Dynamic Programming.
• Lecture 3: Reinforcement Learning Algorithms. Notebook.
• Practical session 1: Dynamic Programming and Q-Learning.
• Lecture 4: Reinforcement Learning with Function Approximation.
• Practical session 2: Deep Q Networks.
• Lecture 5: Beyond Value-Based Methods.
• Lecture 6: Multi-armed bandits.
• Practical session 3: Bandit algorithms.
• Lecture 7: Bandit tools for RL.

List of possible projects (you are welcome to choose your own and e-mail me).

References

• Bandit Algorithms. Tor Lattimore and Csaba Szepesvari (2019).
• Reinforcement Learning. Richard Sutton and Andrew Barto (2018 edition).
• Reinforcement Learning Algorithms. Csaba Szepesvari (2009).
• Markov Decision Processes. Martin Puterman (1994).
• Nice lecture notes written by several colleagues: Rémi Munos, Alessandro Lazaric and Aurélien Garivier.
• Material from the 1st Reinforcement Learning Summer School.