Markov Decision Models with Weighted Discounted Criteria

Simple item page
dc.contributor.authorFeinberg, Eugene A.en_US
dc.contributor.authorShwartz, Adamen_US
dc.contributor.departmentISRen_US
dc.date.accessioned2007-05-23T09:47:55Z
dc.date.available2007-05-23T09:47:55Z
dc.date.issued1991en_US
dc.description.abstractWe consider a discrete time Markov Decision Process with infinite horizon. The criterion to be maximized is the sum of a number of standard discounted rewards, each with a different discount factor. Situations in which such criteria arise include modeling investments, modeling projects of different durations and systems with different time-scales, and some axiomatic formulations of multi-attribute preference theory. We show that for this criterion for some positive e there need not exist an e - optimal (randomized) stationary strategy, even when the state and action sets are finite. However, e - optimal Markov (non-randomized) strategies and optimal Markov strategies exist under weak conditions. We exhibit e - optimal Markov strategies which are stationary from some time onward. When both state and action spaces are finite, there exists an optimal Markov strategy with this property. We provide an explicit algorithm for the computation of such strategies.en_US
dc.format.extent763517 bytes
dc.format.mimetypeapplication/pdf
dc.identifier.urihttp://hdl.handle.net/1903/5091
dc.language.isoen_USen_US
dc.relation.ispartofseriesISR; TR 1991-43en_US
dc.subjectqueuing networksen_US
dc.subjectMarkov optimizationen_US
dc.subjectPerformanceen_US
dc.subjectcommunication networksen_US
dc.titleMarkov Decision Models with Weighted Discounted Criteriaen_US
dc.typeTechnical Reporten_US

Files

Original bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
TR_91-43.pdf
Size:
745.62 KB
Format:
Adobe Portable Document Format
Download

Collections

Institute for Systems Research Technical Reports