Teaching · Lecture
Introduction to Reinforcement Learning
Classic and modern reinforcement learning: how agents learn to make good decisions from interaction, from Markov decision processes and dynamic programming to deep actor-critic methods.
Lecture
Summer semester
In English
When
Every summer semester at the University of Würzburg, taught in English.
Course book
R. S. Sutton and A. G. Barto, Reinforcement Learning: An Introduction.
Tools
Algorithms are implemented in Python. Recommended library: MushroomRL, developed in our group.
Lectures
- Learning agents
- Markov decision processes
- Dynamic programming
- Multi-armed bandits & exploration vs. exploitation
- Model-free estimation
- Model-free control
- RL with value function approximation
- Introduction to deep RL
- Policy gradient
- Introduction to actor-critic approaches
- Deep actor-critic algorithms
- Model-based RL and Monte-Carlo tree search
Instructors
Sebastian Griesbach
Teaching assistant
sebastian.griesbach@uni-wuerzburg.de
Mahdi Kallel
Teaching assistant
mahdi.kallel@uni-wuerzburg.de
Readings
- Course book: R. S. Sutton and A. G. Barto. Reinforcement Learning: An Introduction. MIT Press, 2018.
- C. Szepesvári. Algorithms for Reinforcement Learning. Synthesis Lectures on AI and ML, 2010.
- M. P. Deisenroth, A. A. Faisal and C. S. Ong. Mathematics for Machine Learning. Cambridge University Press, 2020.