Finite Memory Policies for Partially Observable Markov Decision Proesses

Finite Memory Policies for Partially Observable Markov Decision Proesses

This dissertation makes contributions to areas of research on planning with POMDPs: complexity theoretic results and heuristic techniques. The most important contributions are probably the complexity of approximating the optimal history-dependent finite-horizon policy for a POMDP, and the idea of he...

Full description

Bibliographic Details
Main Author: Lusena, Christopher
Format: Others
Published: UKnowledge 2001
Subjects:
Partially Observable Decision Processes|Markov Decision|Processes|POMPP|Heuristics|Complexity Theory
Online Access:http://uknowledge.uky.edu/gradschool_diss/323
http://uknowledge.uky.edu/cgi/viewcontent.cgi?article=1326&context=gradschool_diss
id ndltd-uky.edu-oai-uknowledge.uky.edu-gradschool_diss-1326
record_format oai_dc
spelling ndltd-uky.edu-oai-uknowledge.uky.edu-gradschool_diss-13262015-04-11T05:01:30Z Finite Memory Policies for Partially Observable Markov Decision Proesses Lusena, Christopher This dissertation makes contributions to areas of research on planning with POMDPs: complexity theoretic results and heuristic techniques. The most important contributions are probably the complexity of approximating the optimal history-dependent finite-horizon policy for a POMDP, and the idea of heuristic search over the space of FFTs. 2001-01-01T08:00:00Z text application/pdf http://uknowledge.uky.edu/gradschool_diss/323 http://uknowledge.uky.edu/cgi/viewcontent.cgi?article=1326&context=gradschool_diss University of Kentucky Doctoral Dissertations UKnowledge Partially Observable Decision Processes|Markov Decision|Processes|POMPP|Heuristics|Complexity Theory
collection NDLTD
format Others
sources NDLTD
topic Partially Observable Decision Processes|Markov Decision|Processes|POMPP|Heuristics|Complexity Theory
spellingShingle Partially Observable Decision Processes|Markov Decision|Processes|POMPP|Heuristics|Complexity Theory
Lusena, Christopher
Finite Memory Policies for Partially Observable Markov Decision Proesses
description This dissertation makes contributions to areas of research on planning with POMDPs: complexity theoretic results and heuristic techniques. The most important contributions are probably the complexity of approximating the optimal history-dependent finite-horizon policy for a POMDP, and the idea of heuristic search over the space of FFTs.
author Lusena, Christopher
author_facet Lusena, Christopher
author_sort Lusena, Christopher
title Finite Memory Policies for Partially Observable Markov Decision Proesses
title_short Finite Memory Policies for Partially Observable Markov Decision Proesses
title_full Finite Memory Policies for Partially Observable Markov Decision Proesses
title_fullStr Finite Memory Policies for Partially Observable Markov Decision Proesses
title_full_unstemmed Finite Memory Policies for Partially Observable Markov Decision Proesses
title_sort finite memory policies for partially observable markov decision proesses
publisher UKnowledge
publishDate 2001
url http://uknowledge.uky.edu/gradschool_diss/323
http://uknowledge.uky.edu/cgi/viewcontent.cgi?article=1326&context=gradschool_diss
work_keys_str_mv AT lusenachristopher finitememorypoliciesforpartiallyobservablemarkovdecisionproesses
_version_ 1716800543294226432

Similar Items