Improving value function approximation in factored POMDPs by exploiting model structure

Conference Paper (2015)
Author(s)

Tiago S. Veiga (Lisbon Technical University)

Matthijs Spaan (TU Delft - Algorithmics)

Pedro U. Lima (Lisbon Technical University)

Research Group
Algorithmics
More Info
expand_more
Publication Year
2015
Language
English
Research Group
Algorithmics
Volume number
3
Pages (from-to)
1827-1828
ISBN (electronic)
9781450337717

Abstract

Linear value function approximation in Markov decision processes (MDPs) has been studied extensively, but there are several challenges when applying such techniques to partially observable MDPs (POMDPs). Furthermore, the system designer often has to choose a set of basis functions. We propose an automatic method to derive a suitable set of basis functions by exploiting the structure of factored models. We experimentally show that our approximation can reduce the solution size by several orders of magnitude in large problems.

No files available

Metadata only record. There are no files for this record.