Planning in Decentralized POMDPs with Predictive Policy Representations
2008
Conference Paper
ei
We discuss the problem of policy representation in stochastic and partially observable systems, and address the case where the policy is a hidden parameter of the planning problem. We propose an adaptation of the Predictive State Representations (PSRs) to this problem by introducing tests (sequences of actions and observations) on policies. The new model, called the Predictive Policy Representations (PPRs), is potentially more compact than the other representations, such as decision trees or Finite-State Controllers (FSCs). In this paper, we show how PPRs can be used to improve the performances of a point-based algorithm for DEC-POMDP.
Author(s): | Boularias, A. and Chaib-Draa, B. |
Book Title: | MASPLAN 2008 |
Journal: | Proceedings of ICAPS 2008 Multiagent Planning Workshop (MASPLAN 2008) |
Pages: | 1-7 |
Year: | 2008 |
Month: | September |
Day: | 0 |
Department(s): | Empirische Inferenz |
Bibtex Type: | Conference Paper (inproceedings) |
Event Name: | ICAPS 2008 Multiagent Planning Workshop |
Event Place: | Sydney, Australia |
Digital: | 0 |
Language: | en |
Organization: | Max-Planck-Gesellschaft |
School: | Biologische Kybernetik |
Links: |
PDF
|
BibTex @inproceedings{6836, title = {Planning in Decentralized POMDPs with Predictive Policy Representations}, author = {Boularias, A. and Chaib-Draa, B.}, journal = {Proceedings of ICAPS 2008 Multiagent Planning Workshop (MASPLAN 2008)}, booktitle = {MASPLAN 2008}, pages = {1-7}, organization = {Max-Planck-Gesellschaft}, school = {Biologische Kybernetik}, month = sep, year = {2008}, doi = {}, month_numeric = {9} } |