RL: Reinforcement Learning

Welcome to Reinforcement Learning

Course Information

Reinforcement Learning (RL) is a 10-credit course at Level 11. It runs in Semester 2. The course mark consists of two parts: 50% of the course mark is based on a programming assignment, which is released in mid-February. The other 50% of the course mark is based on the exam, which is in April/May. The University descriptor is here.

Learning Outcomes

On successful completion of this course, you should be able to: 

1. Gain knowledge of basic and advanced reinforcement learning techniques.
2. Identify suitable learning tasks to which these learning techniques can be applied.
3. Appreciate the current limitations of reinforcement learning techniques.
4. Formulate decision problems, set up and run computational experiments, and evaluation of results from experiments.

Main topics to be covered include the following: 

• Reinforcement learning framework
• Bandit problems and action selection
• Dynamic programming
• Monte Carlo methods
• Temporal-difference learning
• Planning in RL
• Function approximation for generalisation
• Actor-critic and gradient-based optimisation
• Multi-agent reinforcement learning
• Training agents and evaluating performance

Relevant QAA Computing Curriculum Sections: Artificial Intelligence, Data Structures and Algorithms, Intelligent Information Systems Technologies, Simulation and Modelling.