Header logo is

State Space Compression with Predictive Representations

2008

Conference Paper

ei


Current studies have demonstrated that the representational power of predictive state representations (PSRs) is at least equal to the one of partially observable Markov decision processes (POMDPs). This is while early steps in planning and generalization with PSRs suggest substantial improvements compared to POMDPs. However, lack of practical algorithms for learning these representations severely restricts their applicability. The computational inefficiency of exact PSR learning methods naturally leads to the exploration of various approximation methods that can provide a good set of core tests through less computational effort. In this paper, we address this problem in an optimization framework. In particular, our approach aims to minimize the potential error that may be caused by missing a number of core tests. We provide analysis of the error caused by this compression and present an empirical evaluation illustrating the performance of this approach.

Author(s): Boularias, A. and Izadi, M. and Chaib-Draa, B.
Book Title: Flairs 2008
Journal: Proceedings of 21st International Florida Artificial Intelligence Research Society Conference (FLAIRS 2008)
Pages: 41-46
Year: 2008
Month: May
Day: 0
Editors: Wilson, D. C., H. C. Lane
Publisher: AAAI Press

Department(s): Empirical Inference
Bibtex Type: Conference Paper (inproceedings)

Event Name: 21st International Florida Artificial Intelligence Research Society Conference
Event Place: Coconut Grove, FL, USA

Address: Menlo Park, CA, USA
Digital: 0
Language: en
Organization: Max-Planck-Gesellschaft
School: Biologische Kybernetik

Links: PDF
Web

BibTex

@inproceedings{6831,
  title = {State Space Compression with Predictive Representations},
  author = {Boularias, A. and Izadi, M. and Chaib-Draa, B.},
  journal = {Proceedings of 21st International Florida Artificial Intelligence Research Society Conference (FLAIRS 2008)},
  booktitle = {Flairs 2008},
  pages = {41-46},
  editors = {Wilson, D. C., H. C. Lane},
  publisher = {AAAI Press},
  organization = {Max-Planck-Gesellschaft},
  school = {Biologische Kybernetik},
  address = {Menlo Park, CA, USA},
  month = may,
  year = {2008},
  month_numeric = {5}
}