Reinforcement learning

Learning Results

Upon successful completion of the course, students will be able to:

• Understand the main features of reinforcement learning that distinguishes it from AI and non-interactive machine learning
• Given an application problem (e.g. from computer vision, robotics, etc), judge if it should be formulated as a RL problem, define it formally and select best-suited algorithm.
• Develop in code common RL algorithms.
• Describe multiple criteria for analyzing RL algorithms and evaluate algorithms

Skills acquired

• Data and information retrieval, analysis and synthesis
• Decision making
• Individual work
• Team work
• New research ideas generation
• Promotion of free, creative and inductive thinking

• Introduction to Reinforcement Learning
• Markov Decision Processes
• Model-free prediction
• Model-free control
• Q-learning
• Value Function Approximation
• Policy Gradient
• Actor-critic and gradient-based optimization
• Multi-agent reinforcement learning
• Deep reinforcement learning

1. Reinforcement Learning: An Introduction, Sutton and Barto, 2nd Edition.
2. Reinforcement Learning: State-of-the-Art, Marco Wiering and Martijn van Otterlo, Eds.

Related scientific journals:

1. Artificial Intelligence
2. IEEE Transactions on Artificial Intelligence

3. NeurIPS conference proceedings