Toward approximate planning in very large stochastic domains

Ann E. Nicholson, Leslie Pack Kaelbling

Research output: Contribution to conferencePaperpeer-review

Abstract

In this paper we extend previous work on approximate planning in large stochastic domains by adding the ability to plan in automatically-generated abstract world views. The dynamics of the domain are represented compositionally using a Bayesian network. Sensitivity analysis is performed on the network to identify the aspects of the world upon which success is most highly dependent. An abstract world model is constructed by including only the most relevant aspects of the world. The world view can be refined over time, making the overall planner behave in most cases like an anytime algorithm. This paper is a preliminary report on this ongoing work.

Original languageEnglish
Pages190-196
Number of pages7
Publication statusPublished - 1994
Externally publishedYes
EventAAAI Spring Symposium 1994 - Palo Alto, United States of America
Duration: 21 Mar 199423 Mar 1994

Conference

ConferenceAAAI Spring Symposium 1994
Country/TerritoryUnited States of America
CityPalo Alto
Period21/03/9423/03/94

Cite this