English

Pure Strategy Best Responses to Mixed Strategies in Repeated Games

Computer Science and Game Theory 2019-02-26 v1

Abstract

Repeated games are difficult to analyze, especially when agents play mixed strategies. We study one-memory strategies in iterated prisoner's dilemma, then generalize the result to k-memory strategies in repeated games. Our result shows that there always exists a pure strategy best response, which can be computed with SMT or MDP solvers. However, there may not exist such pure strategy best response in multi-agent tournaments. All source code is released for verification.

Keywords

Cite

@article{arxiv.1902.09066,
  title  = {Pure Strategy Best Responses to Mixed Strategies in Repeated Games},
  author = {Shiheng Wang and Fangzhen Lin},
  journal= {arXiv preprint arXiv:1902.09066},
  year   = {2019}
}

Comments

6 pages, 1 reference page. 1 figure