Back to Search
Start Over
Improved POMDP Tree Search Planning with Prioritized Action Branching
- Source :
- AAAI-21 Technical Tracks Vol. 35, No. 13, 2021, 11888-11894
- Publication Year :
- 2020
-
Abstract
- Online solvers for partially observable Markov decision processes have difficulty scaling to problems with large action spaces. This paper proposes a method called PA-POMCPOW to sample a subset of the action space that provides varying mixtures of exploitation and exploration for inclusion in a search tree. The proposed method first evaluates the action space according to a score function that is a linear combination of expected reward and expected information gain. The actions with the highest score are then added to the search tree during tree expansion. Experiments show that PA-POMCPOW is able to outperform existing state-of-the-art solvers on problems with large discrete action spaces.<br />Comment: 7 pages
Details
- Database :
- arXiv
- Journal :
- AAAI-21 Technical Tracks Vol. 35, No. 13, 2021, 11888-11894
- Publication Type :
- Report
- Accession number :
- edsarx.2010.03599
- Document Type :
- Working Paper