The lecture gives an introduction to the basics of reinforcement learning and covers
the following topics:
- Introduction: Definition of the Reinforcement Learning Problem
- Introduction: Scientific Computing with Python
- Multi-arm Bandits
- Finite Markov Decision Processes
- Dynamic Programming
- Monte Carlo Methods
- Temporal-Difference Learning
- Policy Gradient Methods
- Deep Reinforcement Learning (A Selection)
- Selected Applications and Examples of Reinforcement Learning
The content of this lecture is mainly based on the book:
"Reinforcement Learning: An Introduction by Richard S. Sutton and Andrew G. Barto.
A draft of the book should be available
here.
There will also be complementary practical exercieses to improve the understanding of the material tought in the class in the form of homework assignments.
The assignments consist of analytical problems and programming exercises. Programming exercises will be solved using the Python programming language (no prior knowledge is required, there will be a short introduction in the first lesson).
Questions, suggestions, complaints, etc.