Generalized Value Bounds and Column Reduction in Finite Markov Decision Problems

R. G. Vickson
Additional contact information
R. G. Vickson: University of Waterloo, Waterloo, Ontario

Operations Research, 1980, vol. 28, issue 2, 387-394

Abstract: Generalized value-function bounds are developed for a class of value-iteration methods in finite state and action Markov decision problems. The bounds are applicable to either discounted total value problems, or to undiscounted total value problems for absorbed processes which are multistage contracting. Some effects of Porteus' column reduction transformation on value-function bounds are studied. It is shown that column reduction tightens the bounds but does not accelerate their asymptotic rate of convergence. Explicit formulas are given for the value bounds obtained through column reduction, thus eliminating the need for performing the transformation.

Date: 1980
References: Add references at CitEc
Citations:

Downloads: (external link)
http://dx.doi.org/10.1287/opre.28.2.387 (application/pdf)

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:inm:oropre:v:28:y:1980:i:2:p:387-394

Access Statistics for this article

More articles in Operations Research from INFORMS Contact information at EDIRC.
Bibliographic data for series maintained by Chris Asher ().