A Learning Algorithm for Risk-Sensitive Cost

被引：34

作者：

Basu, Arnab ^{[1
]}

Bhattacharyya, Tirthankar ^{[2
]}

Borkar, Vivek S. ^{[3
]}

机构：

[1] Indian Inst Management Bangalore, Quantitat Methods & Informat Syst Area, Bangalore 560076, Karnataka, India

[2] Indian Inst Sci, Dept Math, Bangalore 560012, Karnataka, India

[3] Tata Inst Fundamental Res, Sch Technol & Comp Sci, Bombay 400005, Maharashtra, India

来源：

MATHEMATICS OF OPERATIONS RESEARCH | 2008年 / 33卷 / 04期

关键词：

learning algorithm; risk-sensitive cost; function approximation; stochastic approximation;

D O I：

10.1287/moor.1080.0324

中图分类号：

C93 [管理学]; O22 [运筹学];

学科分类号：

070105 ; 12 ; 1201 ; 1202 ; 120202 ;

摘要：

A linear function approximation-based reinforcement learning algorithm is proposed for Markov decision processes with infinite horizon risk-sensitive cost. Its convergence is proved using the "o.d.e. method" for stochastic approximation. The scheme is also extended to continuous state space processes.

引用

页码：880 / 898

页数：19

共 50 条

[1] A Policy Gradient Algorithm for the Risk-Sensitive Exponential Cost MDP
Moharrami, Mehrdad
Murthy, Yashaswini
Roy, Arghyadip
Srikant, R.
MATHEMATICS OF OPERATIONS RESEARCH, 2025, 50 (01)
[2] Risk-Sensitive Reinforcement Learning
Shen, Yun
Tobia, Michael J.
Sommer, Tobias
Obermayer, Klaus
NEURAL COMPUTATION, 2014, 26 (07) : 1298 - 1328
[3] Learning Bounds for Risk-sensitive Learning
Lee, Jaeho
Park, Sejun
Shin, Jinwoo
ADVANCES IN NEURAL INFORMATION PROCESSING SYSTEMS 33, NEURIPS 2020, 2020, 33
[4] Risk-sensitive reinforcement learning
Mihatsch, O
Neuneier, R
MACHINE LEARNING, 2002, 49 (2-3) : 267 - 290
[5] Risk-Sensitive Reinforcement Learning
Oliver Mihatsch
Ralph Neuneier
Machine Learning, 2002, 49 : 267 - 290
[6] A sensitivity formula for risk-sensitive cost and the actor-critic algorithm
Borkar, VS
SYSTEMS & CONTROL LETTERS, 2001, 44 (05) : 339 - 346
[7] Risk-sensitive online learning
Even-Dar, Eyal
Kearns, Michael
Wortman, Jennifer
ALGORITHMIC LEARNING THEORY, PROCEEDINGS, 2006, 4264 : 199 - 213
[8] Inverse Risk-Sensitive Reinforcement Learning
Ratliff, Lillian J.
Mazumdar, Eric
IEEE TRANSACTIONS ON AUTOMATIC CONTROL, 2020, 65 (03) : 1256 - 1263
[9] Risk-Sensitive Control with Near Monotone Cost
Biswas, Anup
Borkar, V. S.
Kumar, K. Suresh
APPLIED MATHEMATICS AND OPTIMIZATION, 2010, 62 (02): : 145 - 163
[10] Exponential TD Learning: A Risk-Sensitive Actor-Critic Reinforcement Learning Algorithm
Noorani, Erfaun
Mavridis, Christos N.
Baras, John S.
2023 AMERICAN CONTROL CONFERENCE, ACC, 2023, : 4104 - 4109

← 1 2 3 4 5 →