Reinforcement Learning 2026HT, 6.0 credits Status Open for interest registrations School IDA-gemensam (IDA) Division STIMA Owner José M Peña

Log in

---

Course plan

No of lectures

5 lectures of 3 hours each.

Recommended for

Students in the field of machine learning, artificial intelligence, or statistics.

The course was last given

In june 2025, as part of the 17th Machine Learning and Advanced Statistics Summer School https://cig.fi.upm.es/mlas/

Goals

Introduction to classical algorithms in reinforcement learning, and implementation of them.

Prerequisites

Basic statistics and probability theory. Basic knowledge of R or Python.

Organization

Lectures, each of which includes some practical exercises in R or Python.

Content

See lectures.

Literature

Sutton, R. S. and Barto, A. G. Reinforcement Learning: An Introduction.
The MIT Press, 2018. http://incompleteideas.net/book/the-book-2nd.html

Lectures

Lecture 0
- Introduction

Lecture 1
- Learning through Interaction
- Value Iteration
- Policy Iteration
- Monte Carlo Control
- Practical Session

Lecture 2
- Q-Learning
- Sarsa and Expected Sarsa
- Practical Session

Lecture 3
- REINFORCE
- Practical Session

Lecture 4
- Deep Q-Learning
- Practical Session

Lecture 5
- Off-line Learning
- Practical Session

Examination

Exercise reports and presentations of results (both in pairs).

Examiner

Jose M. Peña

Credits

6 hp

Comments

Page responsible: Anne Moe