An average-value-at-risk criterion for Markov decision processes with unbounded costs  

作　　者：Qiuli LIU Wai-Ki CHING Junyu ZHANG Hongchu WANG 

机构地区：[1]School of Mathematical Sciences,South China Normal University,Guangzhou,510631,China [2]Advanced Modeling and Applied Computing Laboratory,Department of Mathematics,The University of Hong Kong,Hong Kong,China [3]School of Mathematics,Sun Yat-Sen University,Guangzhou,510275,China

出　　处：《Frontiers of Mathematics in China》2022年第4期673-687,共15页中国高等学校学术文摘·数学（英文）

基　　金：supported by the National Natural Science Foundation of China(Grant Nos.61673019,11931018);the Natural Science Foundation of Guangdong Province(Grant Nos.2018A030313738,2021A1515010057);Guangdong Province Key Laboratory of Computational Science at the Sun Yat-sen University(2020B1212060032);IMR and RAE Research Fund,Faculty of Science,HKU.

摘　　要：We study the Markov decision processes under the average-value-at-risk criterion.The state space and the action space are Borel spaces,the costs are admitted to be unbounded from above,and the discount factors are state-action dependent.Under suitable conditions,we establish the existence of optimal deterministic stationary policies.Furthermore,we apply our main results to a cash-balance model.

关 键 词：Markov decision processes average-value-at-risk(AVaR) state-action dependent discount factors optimal policy 

分 类 号：O17[理学—数学]