Title
Dynamic Non-uniform Abstractions for Approximate Planning in Large Structured Stochastic Domains
Abstract
ions forApproximate Planning in Large StructuredStochastic DomainsJ. Baum and A.E. NicholsonSchool of Computer Science and Software EngineeringMonash University, Clayton, VICTORIA 3168, Australiaphone: +61 3 9905-5225 fax: +61 3 9905-5146fjirib,annng@cs.monash.edu.auAbstract. The theory of Markov Decision Processes (MDPs) providesalgorithms for generating an optimal policy. For large domains these algorithmsbecome intractable and approximate solutions become necessary.In this paper ...
Year
DOI
Venue
1998
10.1007/BFb0095303
PRICAI
Keywords
Field
DocType
large structured stochastic domains,approximate planning,dynamic non-uniform abstractions,markov decision process
Dynamic programming,Abstraction,Cross product,Computer science,Markov decision process,Exploit,Real-time operating system,Artificial intelligence,Stochastic programming,State space,Machine learning
Conference
Volume
ISSN
ISBN
1531
0302-9743
3-540-65271-X
Citations 
PageRank 
References 
7
0.90
11
Authors
2
Name
Order
Citations
PageRank
Jiri Baum170.90
Ann E. Nicholson269288.01