Non-stationary Bandits with Heavy Tail

被引：0
|
作者：
Pan, Weici [1 ]
Liu, Zhenhua [1 ]
机构：
[1] Stony Brook University, United States
来源：
Performance Evaluation Review | 2024年 / 52卷 / 02期
关键词：
Gaussian assumption - Heavy-tailed - Heavy-tails - Multiarmed bandits (MABs) - Nonstationary - Performance - Risk neutrals - Sub-Gaussians;
D O I：
10.1145/3695411.3695424
中图分类号：
学科分类号：
摘要：
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. © 2024 Copyright is held by the owner/author(s).
引用
收藏
页码：33 / 35
相关论文
共 50 条
  • [21] A New Look at Dynamic Regret for Non-Stationary Stochastic Bandits
    Abbasi-Yadkori, Yasin
    Gyorgy, Andraes
    Lazic, Nevena
    JOURNAL OF MACHINE LEARNING RESEARCH, 2023, 24
  • [22] Stochastic Bandits With Non-Stationary Rewards: Reward Attack and Defense
    Yang, Chenye
    Liu, Guanlin
    Lai, Lifeng
    IEEE TRANSACTIONS ON SIGNAL PROCESSING, 2024, 72 : 5007 - 5020
  • [23] Optimal and Efficient Dynamic Regret Algorithms for Non-Stationary Dueling Bandits
    Saha, Aadirupa
    Gupta, Shubham
    INTERNATIONAL CONFERENCE ON MACHINE LEARNING, VOL 162, 2022, : 19027 - 19049
  • [24] Non-stationary Continuum-armed Bandits for Online Hyperparameter Optimization
    Lu, Shiyin
    Zhou, Yu-Hang
    Shi, Jing-Cheng
    Zhu, Wenya
    Yu, Qingtao
    Chen, Qing-Guo
    Da, Qing
    Zhang, Lijun
    WSDM'22: PROCEEDINGS OF THE FIFTEENTH ACM INTERNATIONAL CONFERENCE ON WEB SEARCH AND DATA MINING, 2022, : 618 - 627
  • [25] Achieving Optimal Dynamic Regret for Non-stationary Bandits without Prior Information
    Auer, Peter
    Chen, Yifang
    Gajane, Pratik
    Lee, Chung-Wei
    Luo, Haipeng
    Ortner, Ronald
    Wei, Chen-Yu
    CONFERENCE ON LEARNING THEORY, VOL 99, 2019, 99
  • [26] ANACONDA: An Improved Dynamic Regret Algorithm for Adaptive Non-Stationary Dueling Bandits
    Buening, Thomas Kleine
    Saha, Aadirupa
    INTERNATIONAL CONFERENCE ON ARTIFICIAL INTELLIGENCE AND STATISTICS, VOL 206, 2023, 206
  • [27] A Change-Detection-Based Thompson Sampling Framework for Non-Stationary Bandits
    Ghatak, Gourab
    IEEE TRANSACTIONS ON COMPUTERS, 2021, 70 (10) : 1670 - 1676
  • [28] Contextual Multi-Armed Bandits for Non-Stationary Wireless Network Selection
    Martinez, Lluis
    Vidal, Josep
    Cabrera-Bean, Margarita
    IEEE CONFERENCE ON GLOBAL COMMUNICATIONS, GLOBECOM, 2023, : 285 - 290
  • [29] Non-Stationary Bandits under Recharging Payoffs: Improved Planning with Sublinear Regret
    Papadigenopoulos, Orestis
    Caramanis, Constantine
    Shakkottai, Sanjay
    ADVANCES IN NEURAL INFORMATION PROCESSING SYSTEMS 35, NEURIPS 2022, 2022,
  • [30] A Technical Note on Non-Stationary Parametric Bandits: Existing Mistakes and Preliminary Solutions
    Faury, Louis
    Russac, Yoan
    Abeille, Marc
    Calauzenes, Clement
    ALGORITHMIC LEARNING THEORY, VOL 132, 2021, 132
12345