Perseus: Randomized Point-based Value Iteration for POMDPs

Details Perseus: Randomized Point-based Value Iteration for POMDPs Perseus: Randomized Point-based Value Iteration for POMDPs

2011-09-09

arxiv.org/abs/1109.2145v1

Journal Of Artificial Intelligence Research, Volume 24, pages 195-220, 2005

Computer Science

Artificial Intelligence

Scientific paper

10.1613/jair.1659

Partially observable Markov decision processes (POMDPs) form an attractive and principled framework for agent planning under uncertainty. Point-based approximate techniques for POMDPs compute a policy based on a finite set of points collected in advance from the agents belief space. We present a randomized point-based value iteration algorithm called Perseus. The algorithm performs approximate value backup stages, ensuring that in each backup stage the value of each point in the belief set is improved; the key observation is that a single backup may improve the value of many belief points. Contrary to other point-based methods, Perseus backs up only a (randomly selected) subset of points in the belief set, sufficient for improving the value of each belief point in the set. We show how the same idea can be extended to dealing with continuous action spaces. Experimental results show the potential of Perseus in large scale POMDP problems.

Affiliated with

Also associated with

No associations

Rating

Perseus: Randomized Point-based Value Iteration for POMDPs does not yet have a rating. At this time, there are no reviews or comments for this scientific paper.

If you have personal experience with Perseus: Randomized Point-based Value Iteration for POMDPs, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Perseus: Randomized Point-based Value Iteration for POMDPs will most certainly appreciate the feedback.

Rate now

Comments { 0 }

Profile ID: LFWR-SCP-O-38101