Reinforcement Learning

Objectives

The goal of this course is to provide students an overview of reinforcement learning, a very debated machine learning subfield. Reinforcement learning is the process of creating algorithms that learn to predict and behave in a stochastic environment based on previous experience. Reinforcement learning has a wide range of applications, from traditional control issues like engine optimization or dynamical system control to game play, inventory control, and a variety of other domains.

General characterization

Code

200295

Credits

4.0

Responsible teacher

Nuno Tiago Falcão Alpalhão

Hours

Weekly - Available soon

Total - Available soon

Teaching language

Portuguese. If there are Erasmus students, classes will be taught in English

Prerequisites

TBD

Bibliography

Teaching method

TBD

Evaluation method

TBD

Subject matter

  1. Introduction to Reinforcement Learning
  2. Sequential Decision-Making
  3. Markov Decision Processes
  4. Value Functions & Bellman Equations
  5. Dynamic Programming