Abstract | ||
---|---|---|
High dimensionality of belief space in partially observable Markov decision processes (POMDPs) is one of the major causes that severely restricts the applicability of this model. Previous studies have demonstrated that the dimensionality of a POMDP can eventually be reduced by transforming it into an equivalent predictive state representation (PSR). In this paper, we address the problem of finding an approximate and compact PSR model corresponding to a given POMDP model. We formulate this problem in an optimization framework. Our algorithm tries to minimize the potential error that missing some core tests may cause. We also present an empirical evaluation on benchmark problems, illustrating the performance of this approach. |
Year | DOI | Venue |
---|---|---|
2008 | 10.1109/ICMLA.2008.115 | San Diego, CA |
Keywords | Field | DocType |
time-series segmentation,bayesian approach,unsupervised scenario,linear gaussian,fundamental problem,segmentation model,prediction-directed compression,computational modeling,minimisation,history,optimization,partially observable markov decision process,predictive models,prediction algorithms,markov processes,markov decision process,multi agent systems,pomdp,mathematical model,compression,approximation algorithms,vectors,decision theory | Markov process,Computer science,Predictive state representation,Markov decision process,Multi-agent system,Artificial intelligence,Decision theory,Approximation algorithm,Mathematical optimization,Pattern recognition,Partially observable Markov decision process,Curse of dimensionality,Machine learning | Conference |
ISBN | Citations | PageRank |
978-0-7695-3495-4 | 0 | 0.34 |
References | Authors | |
6 | 3 |
Name | Order | Citations | PageRank |
---|---|---|---|
Boularias, Abdeslam | 1 | 105 | 20.64 |
Masoumeh Izadi | 2 | 5 | 2.00 |
Chaib-draa, Brahim | 3 | 1190 | 113.23 |