ResearchSpace

Belief-node Condensation for Online POMDP Algorithms

Show simple item record

dc.contributor.author Rens, G
dc.contributor.author Ferrein, A
dc.date.accessioned 2014-02-13T08:45:11Z
dc.date.available 2014-02-13T08:45:11Z
dc.date.issued 2013-09
dc.identifier.citation Rens, G and Ferrein, A. 2013. Belief-node Condensation for Online POMDP Algorithms. In: IEEE AFRICON 2013, Mauritius, 9-12 September 2013 en_US
dc.identifier.uri http://www.cair.za.net/sites/default/files/outputs/Rens-AFRICON-13.pdf
dc.identifier.uri http://hdl.handle.net/10204/7190
dc.description IEEE AFRICON 2013, Mauritius, 9-12 September 2013. en_US
dc.description.abstract We consider online partially observable Markov decision processes (POMDPs) which compute policies by local look-ahead from the current belief-state. One problem is that belief-nodes deeper in the decision-tree increase in the number of states with non-zero probability they contain. Computation time of updating a belief-state is exponential in the number of states contained by the belief. Belief-update occurs for each node in a search tree. It would thus pay to reduce the size of the nodes while keeping the information they contain. In this paper, we compare four fast and frugal methods to reduce the size of belief-nodes in the search tree, hence improving the running-time of online POMDP algorithms. en_US
dc.language.iso en en_US
dc.publisher Centre for Artificial Intelligence Research en_US
dc.relation.ispartofseries Workflow;12077
dc.subject Partially observable Markov decision processes en_US
dc.subject POMDPs en_US
dc.title Belief-node Condensation for Online POMDP Algorithms en_US
dc.type Conference Presentation en_US
dc.identifier.apacitation Rens, G., & Ferrein, A. (2013). Belief-node Condensation for Online POMDP Algorithms. Centre for Artificial Intelligence Research. http://hdl.handle.net/10204/7190 en_ZA
dc.identifier.chicagocitation Rens, G, and A Ferrein. "Belief-node Condensation for Online POMDP Algorithms." (2013): http://hdl.handle.net/10204/7190 en_ZA
dc.identifier.vancouvercitation Rens G, Ferrein A, Belief-node Condensation for Online POMDP Algorithms; Centre for Artificial Intelligence Research; 2013. http://hdl.handle.net/10204/7190 . en_ZA
dc.identifier.ris TY - Conference Presentation AU - Rens, G AU - Ferrein, A AB - We consider online partially observable Markov decision processes (POMDPs) which compute policies by local look-ahead from the current belief-state. One problem is that belief-nodes deeper in the decision-tree increase in the number of states with non-zero probability they contain. Computation time of updating a belief-state is exponential in the number of states contained by the belief. Belief-update occurs for each node in a search tree. It would thus pay to reduce the size of the nodes while keeping the information they contain. In this paper, we compare four fast and frugal methods to reduce the size of belief-nodes in the search tree, hence improving the running-time of online POMDP algorithms. DA - 2013-09 DB - ResearchSpace DP - CSIR KW - Partially observable Markov decision processes KW - POMDPs LK - https://researchspace.csir.co.za PY - 2013 T1 - Belief-node Condensation for Online POMDP Algorithms TI - Belief-node Condensation for Online POMDP Algorithms UR - http://hdl.handle.net/10204/7190 ER - en_ZA


Files in this item

Thumbnail
Name: Rens_2013.pdf
Size: 223.6Kb
Format: PDF
View/Open

This item appears in the following Collection(s)

Show simple item record