Title :
Optimality equations in undiscounted Markov decision processes
Author :
Puterman, Martin L.
Author_Institution :
Fac. of Commerce & Bus. Adm., British Columbia Univ., Vancouver, BC, Canada
fDate :
6/21/1905 12:00:00 AM
Abstract :
We explore properties of the average and bias optimality equations in unichain Markov decision processes. We show that in unichain models, these equations have the same form, so that theory for gain optimality carries over to bias optimality
Keywords :
Markov processes; decision theory; matrix algebra; probability; bias optimality; gain optimality; optimality equations; undiscounted Markov decision processes; unichain Markov decision processes; Business; Equations; Fasteners; History; Infinite horizon; Random variables; State-space methods; Stochastic processes; US Department of Energy;
Conference_Titel :
Decision and Control, 1999. Proceedings of the 38th IEEE Conference on
Print_ISBN :
0-7803-5250-5
DOI :
10.1109/CDC.1999.832928
