Department of Information and Computing Sciences

Departement Informatica Onderwijs
Bachelor Informatica Informatiekunde Kunstmatige intelligentie Master Computing Science Game&Media Technology Artifical Intelligence Human Computer Interaction Business Informatics

Onderwijs Informatica en Informatiekunde

Vak-informatie Informatica en Informatiekunde

AI for game technology

Course code:INFOMAIGT
Credits:7.5 ECTS
History:This course was formerly known as Games and agents (INFOMGMAG). You can only do one of these courses.
Period:period 4 (week 17 through 26, i.e., 20-4-2020 through 26-6-2020; retake week 28)
Participants:up till now 45 subscriptions
Schedule:Official schedule representation can be found in MyTimetable
innovatie          Till Miltzow
Jordi Vermeulen
lecture          Till Miltzow
Jordi Vermeulen

In this course, we look at two ways to use AI for games: training an agent to play a game, and automatically generating (elements of) a game.

The first half of the course will focus on using reinforcement learning to train an agent to play a turn-based game. The second half of the course will focus on different methods of procedural content generation.

  • Reinforcement Learning - Suttan and Barto. Available for free online.
  • Procedural Content Generation - Shaker, Togelius and Nelson. Available for free online.
In addition, a selection of papers will be made available.
Course form:Video lectures and two weekly remote interactive sessions.
Exam form:

There will be two homework assignments with theory exercises (H1, H2), and two implementation projects (P1, P2). The final grade will be a weighted sum of the individual grades:

0.25 * H1 + 0.1 * H2 + 0.25 * P1 + 0.4 * P2

The homework exercises will be (partially) peer-graded.

Minimum effort to qualify for 2nd chance exam:To qualify for the retake exam, the grade of the original must be at least 4.