Bounded Policy Iteration for Decentralized POMDPs
Proceedings of the Nineteenth International Joint Conference on Artificial Intelligence. Pages 1287-1291.
Abstract
We present a bounded policy iteration algorithm for infinite-horizon decentralized POMDPs. Policies are represented as joint stochastic finite-state controllers, which consist of a local controller for each agent. We also let a joint controller include a correlation device that allows the agents to correlate their behavior without exchanging information during execution, and show that this leads to improved performance. The algorithm uses a fixed amount of memory, and each iteration is guaranteed to produce a controller with value at least as high as the previous one for all possible initial state distributions. For the case of a single agent, the algorithm reduces to Poupart and Boutilier’s bounded policy iteration for POMDPs.
@inproceedings{ijcai2005p18,
title = {Bounded Policy Iteration for Decentralized POMDPs},
author = {Bernstein, Daniel S. and Hansen, Eric A. and Zilberstein, Shlomo},
booktitle = {Proceedings of the Nineteenth International Joint Conference on
Artificial Intelligence, {IJCAI-05}},
publisher = {International Joint Conferences on Artificial Intelligence},
pages = {1287--1291},
year = {2005},
url = {https://ijcai.org/proceedings/2005/18},
}
TY - CPAPER TI - Bounded Policy Iteration for Decentralized POMDPs AU - Bernstein, Daniel S. AU - Hansen, Eric A. AU - Zilberstein, Shlomo T2 - Proceedings of the Nineteenth International Joint Conference on Artificial Intelligence, IJCAI-05 PB - International Joint Conferences on Artificial Intelligence PY - 2005 SP - 1287 EP - 1291 UR - https://ijcai.org/proceedings/2005/18 ER -Copied
Machine-readable versions of this paper, for tools and AI agents:
