Model Acquisition for Markov Decision Problems
Abstract
The original focus of this work was on the automatic acquisition (learning) of stochastic models. The motivation was the lack of such models for military problems, specifically air-campaign planning, and the existence of new algorithms that could, if the appropriate models were available, considerably improve the accuracy and efficiency of military planning. This final report describes the course of our investigations, some unanticipated turns, and the direction that our research has taken as a consequence of what we have learned.
Document Details
- Document Type
- Technical Report
- Publication Date
- Oct 31, 1998
- Accession Number
- ADA380049
Entities
People
- Thomas Dean
Organizations
- Brown University