Point-Based POMDP Algorithms: Improved Analysis and Implementation

Trey Smith,Reid Simmons
DOI: https://doi.org/10.48550/arXiv.1207.1412
2012-07-05
Abstract:Existing complexity bounds for point-based POMDP value iteration algorithms focus either on the curse of dimensionality or the curse of history. We derive a new bound that relies on both and uses the concept of discounted reachability; our conclusions may help guide future algorithm design. We also discuss recent improvements to our (point-based) heuristic search value iteration algorithm. Our new implementation calculates tighter initial bounds, avoids solving linear programs, and makes more effective use of sparsity.
Artificial Intelligence
What problem does this paper attempt to address?