Minimax Policy for Heavy-Tailed Bandits

被引：3

作者：

Wei, Lai ^{[1
]}

Srivastava, Vaibhav ^{[1
]}

机构：

[1] Michigan State Univ, Dept Elect & Comp Engn, E Lansing, MI 48823 USA

来源：

IEEE CONTROL SYSTEMS LETTERS | 2021年 / 5卷 / 04期

关键词：

Heavy-tailed distribution; stochastic MAB; worst-case regret; minimax policy;

D O I：

10.1109/LCSYS.2020.3035767

中图分类号：

TP [自动化技术、计算机技术];

学科分类号：

0812 ;

摘要：

We study the stochastic Multi-Armed Bandit (MAB) problem under worst-case regret and heavytailed reward distribution. We modify the minimax policy MOSS for the sub-Gaussian reward distribution by using saturated empirical mean to design a new algorithm called Robust MOSS. We show that if the moment of order 1 + epsilon for the reward distribution exists, then the refined strategy has a worst-case regret matching the lower bound while maintaining a distribution-dependent logarithm regret.

引用

页码：1423 / 1428

页数：6

共 50 条

[41] HEAVY-TAILED DISTRIBUTIONS - PROPERTIES AND TESTS
BRYSON, MC
TECHNOMETRICS, 1974, 16 (01) : 61 - 68
[42] The maximum of the periodogram for a heavy-tailed sequence
Mikosch, T
Resnick, S
Samorodnitsky, G
ANNALS OF PROBABILITY, 2000, 28 (02): : 885 - 908
[43] On learning mixtures of heavy-tailed distributions
Dasgupta, A
Hopcroft, J
Kleinberg, J
Sandler, M
46th Annual IEEE Symposium on Foundations of Computer Science, Proceedings, 2005, : 491 - 500
[44] On the emergence of heavy-tailed streamflow distributions
Basso, S.
Schirmer, M.
Botter, G.
ADVANCES IN WATER RESOURCES, 2015, 82 : 98 - 105
[45] Scalar quantisation of heavy-tailed signals
Tsakalides, P
Reveliotis, P
Nikias, CL
IEE PROCEEDINGS-VISION IMAGE AND SIGNAL PROCESSING, 2000, 147 (05): : 475 - 484
[46] Stochastic PDEs with heavy-tailed noise
Chong, Carsten
STOCHASTIC PROCESSES AND THEIR APPLICATIONS, 2017, 127 (07) : 2262 - 2280
[47] Heavy-tailed fractional Pearson diffusions
Leonenko, N. N.
Papic, I.
Sikorskii, A.
Suvak, N.
STOCHASTIC PROCESSES AND THEIR APPLICATIONS, 2017, 127 (11) : 3512 - 3535
[48] On Fuzzy Clustering for Heavy-Tailed Data
Taheri, S. Mahmoud
Mohammadpour, A.
Atiyah, Israa
2017 5TH IRANIAN JOINT CONGRESS ON FUZZY AND INTELLIGENT SYSTEMS (CFIS), 2017, : 202 - 206
[49] Detection of Dependent Heavy-Tailed Signals
Subramanian, Arun
Sundaresan, Ashok
Varshney, Pramod K.
IEEE TRANSACTIONS ON SIGNAL PROCESSING, 2015, 63 (11) : 2790 - 2803
[50] Taming heavy-tailed features by shrinkage
Zhu, Ziwei
Zhou, Wenjing
24TH INTERNATIONAL CONFERENCE ON ARTIFICIAL INTELLIGENCE AND STATISTICS (AISTATS), 2021, 130

← 1 2 3 4 5 →