This course provides in-depth knowledge of the reinforcement learning paradigm. The covodered topics include: 

• Fundamentals of Markov decision problems and dynamic programming
• Monte Carlo methods
• Model-based and model-free reinforcement learning
• Temporal difference learning and Q-learning
• Deep Q-learning
• Policy gradient methods
• Multi-Agent Reinforcement Learning
• Deep Reinforcement Learning for Optimization and Planning

As part of the exercises, students will learn to deepen and apply the topics in a practical way. Selected methods will be implemented in Python using PyTorch and OpenAI Gym.