Game AI middag

Op 30 juni vindt een Game AI middag plaats. Hierbij zullen lezingen gegeven worden door Sander Bakkes (UvT) en Joost Westra (UU), gevolgd door een borrel!

De abstracts luiden als volgt:

Rapid and Reliable Adaptation of Video Game AI
Sander Bakkes, Ph.D.

Current approaches to adaptive game AI typically require numerous trials to learn effective behaviour (i.e., game adaptation is not rapid). In addition, game developers are concerned that applying adaptive game AI may result in uncontrollable and unpredictable behaviour (i.e., game adaptation is not reliable). These characteristics hamper the incorporation of adaptive game AI in commercially available video games. In this presentation, we discuss an alternative to these current approaches. Our alternative approach to adaptive game AI has as its goal adapting rapidly and reliably to game circumstances. Our approach can be classified in the area of case-based adaptive game AI. In the approach, domain knowledge required to adapt to game circumstances is gathered automatically by the game AI, and is exploited immediately (i.e., without trials and without resource-intensive learning) to evoke effective behaviour in a controlled manner in online play. We performed experiments that test case-based adaptive game AI on three different maps in a commercial RTS game. From our results we may conclude that case-based adaptive game AI provides a strong basis for effectively adapting game AI in video games.

Games and Agents
Joost Westra, Ph.D.

Joost Westra is a PhD student at Universiteit Utrecht. His subject is
adapting computer games to the user. Joost has experience with agent
technology in different game engines (Quest 3D, Unreal Tournament,
Quake III and Boswars). He implemented agents that used genetic
algorithms in Quake III for his master thesis. He has given lectures
and supervised the practical assignments in the course “Games and
Agents” at Universiteit Utrecht for four years and has given guest
lectures at Maastricht University on the subject. He also supervised
multiple master students during their thesis work on combining agents
and games.

He will present some general guidelines on coupling agent systems with
game engines. He will also present a quick overview of his master
thesis on "Evolutionary Neural Networks for NPC’s in Quake III." And
he will also talk about his PhD research on adapting games to the
users to optimize learning, while maintaining the storyline.