Non-stationary Bandits with Heavy Tail

Weici Pan; Zhenhua Liu

doi:10.1145/3695411.3695424

Non-stationary Bandits with Heavy Tail

Weici Pan
, Zhenhua Liu

Applied Mathematics and Statistics

Stony Brook University

Research output: Contribution to journal › Article › peer-review

Abstract

In this study, we investigate the performance of multi-armed bandit algorithms in environments characterized by heavytailed and non-stationary reward distributions, a setting that deviates from the conventional risk-neutral and sub- Gaussian assumptions.

Original language	English
Pages (from-to)	33-35
Number of pages	3
Journal	Performance Evaluation Review
Volume	52
Issue number	2
DOIs	https://doi.org/10.1145/3695411.3695424
State	Published - Sep 6 2024

Access to Document

10.1145/3695411.3695424

Cite this

@article{f73a421e88f54c2eaf82ac40da8a3363,

title = "Non-stationary Bandits with Heavy Tail",

abstract = "In this study, we investigate the performance of multi-armed bandit algorithms in environments characterized by heavytailed and non-stationary reward distributions, a setting that deviates from the conventional risk-neutral and sub- Gaussian assumptions.",

author = "Weici Pan and Zhenhua Liu",

note = "Publisher Copyright: {\textcopyright} 2024 Copyright is held by the owner/author(s).",

year = "2024",

month = sep,

day = "6",

doi = "10.1145/3695411.3695424",

language = "English",

volume = "52",

pages = "33--35",

journal = "Performance Evaluation Review",

issn = "0163-5999",

number = "2",

}

TY - JOUR

T1 - Non-stationary Bandits with Heavy Tail

AU - Pan, Weici

AU - Liu, Zhenhua

PY - 2024/9/6

Y1 - 2024/9/6

N2 - In this study, we investigate the performance of multi-armed bandit algorithms in environments characterized by heavytailed and non-stationary reward distributions, a setting that deviates from the conventional risk-neutral and sub- Gaussian assumptions.

AB - In this study, we investigate the performance of multi-armed bandit algorithms in environments characterized by heavytailed and non-stationary reward distributions, a setting that deviates from the conventional risk-neutral and sub- Gaussian assumptions.

UR - https://www.scopus.com/pages/publications/85203527154

U2 - 10.1145/3695411.3695424

DO - 10.1145/3695411.3695424

M3 - Article

AN - SCOPUS:85203527154

SN - 0163-5999

VL - 52

SP - 33

EP - 35

JO - Performance Evaluation Review

JF - Performance Evaluation Review

IS - 2

ER -

Non-stationary Bandits with Heavy Tail

Abstract

Access to Document

Other files and links

Fingerprint

Cite this