Focused Dynamic Programming: Extensive Comparative Results
Abstract
We present a heuristic-based propagation algorithm for solving restricted Markov decision processes (MDPs). Our approach, which combines ideas from deterministic search and recent dynamic programming methods, focuses computation towards promising areas of the state space. It is thus able to significantly reduce the amount of processing required in producing a solution. We present a number of results comparing our approach to existing algorithms on a robotic path planning domain.
Document Details
- Document Type
- Technical Report
- Publication Date
- Mar 01, 2004
- Accession Number
- ADA528892
Entities
People
- Anthony Stentz
- Dave Ferguson
Organizations
- Carnegie Mellon University