Research

Research Interests

My research focuses on Multi-Armed Bandit problems. Generally, I am interested in reinforcement learning and its applications.

Papers

Note: * indicates alphabetic ordering.

  • Variance-Dependent Best Arm Identification (abstract, link)
    Pinyan Lu*, Chao Tao*, Xiaojin Zhang*
    UAI 2021
  • Near-Optimal MNL Bandits Under Risk Criteria (abstract, link)
    Guangyu Xi, Chao Tao, Yuan Zhou
    AAAI 2021
  • Thresholding Bandit with Optimal Aggregate Regret (abstract, link)
    Chao Tao, Saúl A. Blanco, Jian Peng, Yuan Zhou
    NeurIPS 2019
  • Collaborative Learning with Limited Interaction: Tight Bounds for Distributed Exploration in Multi-Armed Bandits (abstract, link)
    Chao Tao*, Qin Zhang*, Yuan Zhou*
    FOCS 2019
  • Best Arm Identification in Linear Bandits with Linear Dimension Dependency (abstract, link)
    Chao Tao, Saúl A. Blanco, Yuan Zhou
    ICML 2018

Talks Given

  • “Near-Optimal MNL Bandits Under Risk Criteria”, Shanghai Jiao Tong University, Remote from Bloomington, IN, 03/07/2021

  • “Near-Optimal MNL Bandits Under Risk Criteria”, Shandong University, Remote from Bloomington, IN, 12/29/2020

  • “Thresholding Bandit with Optimal Aggregate Regret”, Shandong University, Qingdao, China, 01/08/2020

  • “Optimal Maximum Gap Estimation in the Multi-armed Bandit,” INFORMS Annual Meeting, Houston, TX, 2017

Reading Group

Softwares

  • banditpylib

    • A lightweight python library for bandit algorithms

Services

  • Reviewer/Sub-reviewer for: ICDE (2021), SIGMOD (2021), SOSA (2021), VLDB (2020), ICML (2022, 2021, 2020), AAAI (2020), BigData (2019), ISAAC (2019, 2018), NeurIPS (2021, 2020, 2019), IJCAI (2019), ITCS (2018), TCS (2018)