Information Set Generation in Partially Observable Games

Abstract

We address the problem of making single-point decisions in large partially observable games, where players interleave observation, deliberation, and action.  We present information set generation as a key operation needed to reason about games in this way.  We show how this operation can be used to implement an existing decision-making algorithm.  We develop a constraint satisfaction algorithm for performing information set generation and show that it scales better than the existing depth-first search approach on multiple non-trivial games.

Cite

Text

Richards and Amir. "Information Set Generation in Partially Observable Games." AAAI Conference on Artificial Intelligence, 2012. doi:10.1609/AAAI.V26I1.8146

Markdown

[Richards and Amir. "Information Set Generation in Partially Observable Games." AAAI Conference on Artificial Intelligence, 2012.](https://mlanthology.org/aaai/2012/richards2012aaai-information/) doi:10.1609/AAAI.V26I1.8146

BibTeX

@inproceedings{richards2012aaai-information,
  title     = {{Information Set Generation in Partially Observable Games}},
  author    = {Richards, Mark and Amir, Eyal},
  booktitle = {AAAI Conference on Artificial Intelligence},
  year      = {2012},
  pages     = {549-555},
  doi       = {10.1609/AAAI.V26I1.8146},
  url       = {https://mlanthology.org/aaai/2012/richards2012aaai-information/}
}