Towards a Human-in-the-Loop System for Authoring Game AI using Behavior Languages

Erica Kleinman1, Spencer Lynn2, Bryan Loyall2, Magy Seif El-Nasr3

In Proceedings of the 18th International Conference on the Foundations of Digital Games. Lisbon, Portugal (April 2023)

As games get more complicated, Artificially Intelligent (AI) agents need to be better developed to understand and replicate complex,
goal-oriented, reactive behaviors. Many existing behavior language approaches do not often make the AI’s behavior transparent and
readable to the human, making it difficult to debug and update the behavioral model. In this workshop paper, we present a prototype
visualization system that uses a tree-based abstraction to present complex agent behavior to a human user, allowing them to
understand and edit the agent’s behavioral logic in a visual manner.

1 University of California, Santa Cruz
2 Charles River Analytics
3 University of California, Santa Cruz

For More Information

To learn more or request a copy of a paper (if available), contact Spencer Lynn, PhD.

Publication is available from ACM Digital Library, FDG Proceedings.

(Please include your name, address, organization, and the paper reference. Requests without this information will not be honored.)