Reinforcement Learning


Starts on 01/03/2023

Ends on 30/06/2023


Clemens Heitzinger,

Content and organization

Reinforcement learning (RL) is the field of artificial intelligence concerned with the development of strategies that an agent uses to maximize its rewards in a random environment, often without models. Applications include robotics, autonomous driving, (stochastic) optimal control, medicine, and games such as Go, chess, Atari 2600, StarCraft, Gran Turismo, and card games at human or superhuman level.

Chapters in this course include bandit problems, Markov decision problems, Bellman equations, dynamic programming, Monte-Carlo learning, temporal-difference learning, tabular methods, function approximation and deep RL, on-policy and off-policy learning, eligibility traces, policy gradients and actor-critic methods, deep RL, distributional RL, convergence, and PAC (probably approximately correct) estimates.

Lectures notes will be available for download.


Postgraduate. Assumes knowledge of linear algebra, basic calculus and basic probability theory.

Course Duration

Two classes (1.5 hours each) per week.

Course Type

Semester Course



Marking Scheme

From 1 (excellent) to 5 (failed).

Participation terms

Registration required, TBA in February 2023. See course link below and/or my homepage.





Modality (online/in person):

Hybrid; online for remote participants.


There will be two tests. Participation in the tutorial part of the class is mandatory, where students present solutions of (theoretical and programming) exercises.

Host Institution
TU Wien

Other short courses

11. 05. 2022 Go

Geometric learning

05. 04. 2022 Go

Computer Graphics

04. 04. 2022 Go

Bayesian Learning

02. 04. 2022 Go

Computer Graphics

31. 03. 2022 Go

Web of Data

27. 03. 2022 Go

Machine Learning

02. 03. 2022 Go

Player Modeling

28. 02. 2022 Go

Player Modeling