Robustness of policies in constrained Markov decision processes

Author

Zadorojniy, Alexander ; Shwartz, Adam

Author_Institution

Intel Israel Ltd., Haifa, Israel

Volume

Issue

fYear

2006

fDate

4/1/2006 12:00:00 AM

Firstpage

635

Lastpage

638

Abstract

We consider the optimization of finite-state, finite-action Markov decision processes (MDPs), under constraints. Cost and constraints are discounted. We introduce a new method for investigating the continuity, and a certain type of robustness, of the optimal cost and the optimal policy under changes in the constraints. This method is also applicable for other cost criteria such as finite horizon and infinite horizon average cost.

Keywords

Markov processes; cost optimal control; robust control; stochastic systems; constrained Markov decision processes; finite horizon average cost; finite-action process; finite-state process; infinite horizon average cost; optimal cost; optimal policy; robustness; Constraint optimization; Cost function; Delay; Energy consumption; Infinite horizon; Linear programming; Power measurement; Robustness; State-space methods; Throughput; Constrained Markov decision process (MDP); Markov decision processes; discounted cost; robustness; sensitivity;

fLanguage

English

Journal_Title

Automatic Control, IEEE Transactions on

Publisher

ieee

ISSN

0018-9286

Type

jour

DOI

10.1109/TAC.2006.872754

Filename

1618838

Link To Document

https://search.isc.ac/dl/search/defaultta.aspx?DTC=49&DC=896337