Modern Game AI Algorithms (WS 2016/17)


Veranstaltungsnummer
042284

Studiengang/-gänge
Master

Vorlesungsverzeichnis

Learnweb-Plattform

Typ
Vorlesung/Übung

Vorlesungssprache
englisch


Veranstaltungszeitplan

Tag Zeit Häufigkeit Datum Raum
Montag 12:00- 14:00 Uhr wöchentlich 24.10.2016- 10.02.2017 Leonardo-Campus 3, LEO 3.219
Mittwoch 16:00- 18:00 Uhr wöchentlich 26.10.2016- 10.02.2017 Leonardo-Campus 3, LEO 3.219

Hinweis

Beschreibung

In a more traditional understanding, Game AI (artificial
intelligence) comprises a number of algorithms and techniques
that deal with modeling non-player characters (NPC) and their
interactions with other NPC, the game world, or human players.

With the emergence of nonlinear gameplay, much larger game
worlds and much richer interactions, Game AI has started to
grow into domains that had previously been handled without
any algorithmic support, e.g., procedural content generation
(PCG), player experience modeling (PEM), and game mining.

New algorithmic developments as Monte Carlo tree search
(MCTS) and behaviour trees open up new possibilities in
terms of opponent or NPC complexity, even under the the
restricted availability of computational power at real-time.
Nowadays, Game AI is more and more seen as a wide range of
algorithmic solutions for different processes in design,
production, and analysis of computer games. Consequently,
this lecture comprises modern algorithms for the following
domains:

- Non-player character AI
- Player Behavior/Experience Modeling
- Procedural content generation
- Balancing and dynamic difficulty adjustment
- Adaptation and learning in games
- Game mining

Target audience:  Master level students

Knowledge prerequisites:
    basic knowledge of algorithms and data structures,
    basic statistics

Dozenten

  • Dr. Mike Preuß (verantwortlich)