On Bellman's Optimality Principle for zs-POSGs

Olivier Buffet, Jilles Dibangoye, Aurélien Delage, Abdallah Saffidine, Vincent Thomas

Abstract

Many non-trivial sequential decision-making problems are efficiently solved by relying on Bellman's optimality principle, i.e., exploiting the fact that sub-problems are nested recursively within the original problem. Here we show how it can apply to (infinite horizon) 2-player zero-sum partially observable stochastic games (zs-POSGs) by (i) taking a central planner's viewpoint, which can only reason on a sufficient statistic called occupancy state, and (ii) turning such problems into zero-sum occupancy Markov games (zs-OMGs). Then, exploiting the Lipschitz-continuity of the value function in occupancy space, one can derive a version of the HSVI algorithm (Heuristic Search Value Iteration) that provably finds an $\epsilon$-Nash equilibrium in finite time.
BibTeX
@misc{buffet2020bellman,
  author = {Olivier Buffet and Jilles Dibangoye and Aurélien Delage and Abdallah Saffidine and Vincent Thomas},
  title = {{On Bellman's Optimality Principle for zs-POSGs}},
  year = {2020},
  eprint = {2006.16395},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2006.16395},
}
Figure