Sebastian Rosengren and Erik Thorsén: An introduction to Reinforcement Learning

Abstract: Reinforcement learning is as many things are in machine learning both the problem and the solution.    The problem: How should an agent act in an uncertain environment? The Solution: Optimal. In this talk we introduce the underlying theory of reinforcement learning: Markov Decision Processes, Bellman equations, and Value Iteration. As well as provide an introduction to more modern approaches such as Q-learning, and Deep Q-learning. The running example will be blackjack.