The main goal of the course is to broaden and deepen the knowledge acquired during the Artificial Intelligence lecture - but specifically in the context of games.
I would like to show how many diverse roles AI can play when applied to the games domain, both from a research and industry perspective.
The lab exercises will mainly require implementing various agents aimed at solving small-to-medium-sized problems.
Covered subjects will include, among others:
- common MCTS enhancements (MAST, RAVE, AMAF)
- introduction to evolutionary algorithms (RHEA)
- video games AI (Behavioral Trees, HTNs, GOB)
- pathfinding (group, dynamic, hierarchical)
- General Game Playing (GDL, Ludii, RBG/RG)
- Procedural Content Generation (search-based, grammars, game rules)
This lecture, in contrast to the standard AI4Games editions, is meant to be **softcore**. Last edition worked "pretty well" in that regard.
