After successful completion of the course, students are able to understand the theory of certain methods in machine learning and to implement corresponding algorithms. Selected timely application problems are solved in this manner.

This time there will be a focus on reinforcement learning (RL) and timely applications such as:

• deep RL,
• policy evaluation,
• reliability,
• explainability,
• RL with function approximation in scientific and medical applications,
• autonomous driving,
• algorithms for computer games (Atari 2600, ALE (arcade learning environment)),
• algorithms for games such as chess and Go,
• algorithms for card games,
• etc.

Theory: convergence and complexity of various algorithms.

Methods of reinforcement learning, supervised, and unsupervised learning.

Time of first meeting will be announced.

Implementation and documentation.

Linear algebra, analysis.  Experience in programming languages such as Julia or Python.