Mintegia: Urtzi Ayesta "Introduction to Reinforcement Learning"
Noiz eta non
3.1 mintegia - Informatika Fakultatea
Paseo Manuel Lardizabal 1. -20018- Donostia / San Sebastián (Gipuzkoa)
Title: From basic concepts to deep Q-networks
Abstract: Reinforcement learning is a body of theory and techniques for optimal sequential decision making. In its basic setting, at each time, an agent selects an action, and as a result, it collects a reward and the system state evolves. The agent observes the new state and decides on the next action, with the objective of maximizing the total accumulated reward. Reinforcement learning has found numerous applications, ranging from online services (ad placement, recommendation systems), game playing (chess, Atari, Go etc.), control, robotics, etc. In this talk we will first introduce the underlying mathematical framework (Markov decision processes) and its solution methods, including dynamic programming, Monte Carlo methods, and temporal-difference learning. We will also cover how these methods can be combined with parametric function approximation, including deep learning, to find good approximate solutions to real world complexity problems.