The University of Cyprus's MSc Artificial Intelligence is part of the Master programmes in Artificial Intelligence 4 Careers in Europe (MAI4CAREU). One of Master's programme's courses, MAI612 - Machine Learning is split up into several lectures. Taught by Vassilis Vassiliades, PhD, the eighteenth lecture of the MAI612 - Machine Learning course focuses on Model-free Reinforcement Learning.

Learning outcomes

The lesson is divided in five parts: Multi-armed bandits, Model-free Prediction, Between MC and TD: Multi-Step TD, Temporal-Difference Learning for Control, and Optimistic Initialization. In this lesson you will learn about:

• The simpler framework of multi-armed bandits and the exploration-exploitation tradeoff
• Model-free prediction to estimate values in an unknown MDP: Monte Carlo, Temporal-Difference (TD) Learning, and Multi-step TD learning
• Model-free control to optimise values in an unknown MDP: SARSA and Q-learning algorithms
• Optimistic initialization of the value function to help exploration