Functional characterization for average cost Markov decision processes with Doeblin's conditions (Q1179357)
From MaRDI portal
 | This is the item page for this Wikibase entity, intended for internal use and editing purposes. Please use this page instead for the normal view: Functional characterization for average cost Markov decision processes with Doeblin's conditions |
scientific article; zbMATH DE number 24479
| Language | Label | Description | Also known as |
|---|---|---|---|
| English | Functional characterization for average cost Markov decision processes with Doeblin's conditions |
scientific article; zbMATH DE number 24479 |
Statements
Functional characterization for average cost Markov decision processes with Doeblin's conditions (English)
0 references
26 June 1992
0 references
In a former paper of the author [SIAM J. Control Optimization 27, No. 2, 296-307 (1989; Zbl 0677.90085)] the existence of optimal stationary policies was shown for average cost Markov decision processes with compact state and action spaces under a Doeblin-condition. In the present paper the pertinent optimality equations and optimality inequalities are derived, and criteria for optimal policies are given. At first the positive recurrent case is considered, then the problem is treated under rather general assumptions.
0 references
average cost Markov decision processes
0 references
compact state and action spaces
0 references
optimality equations
0 references
optimality inequalities
0 references
criteria for optimal policies
0 references
0 references
0.92160296
0 references
0.8976484
0 references
0.8926091
0 references
0.88863295
0 references
0 references
0.8827301
0 references
0.88179475
0 references
0.87721246
0 references
