CV

Academic background, experience, honors, and research interests.

Contact Information

Name Letian Li
Professional Title Postdoctoral Research Fellow in Statistics and Biostatistics
Email letianli1998@gmail.com
Location Boston, MA
Website https://letianli059.github.io

Professional Summary

Postdoctoral Research Fellow at Harvard Medical School and the Harvard Pilgrim Health Care Institute. My research focuses on reinforcement learning, multi-armed bandits, dynamic pricing, dynamic treatment regimes, heterogeneous effects, and large language models.

Experience

  • Jan 2026 - Present

    Boston, MA, USA

    Postdoctoral Research Fellow
    Harvard Medical School
  • Jan 2026 - Present

    Boston, MA, USA

    Postdoctoral Research Fellow
    Harvard Pilgrim Health Care Institute
  • Jan 2019 - Jun 2019

    Jinan, Shandong, China

    Research Intern
    Zhongtai Securities Co., Ltd.
    Quantitative and Statistical Methods

Education

  • 2020 - 2025

    Hefei, Anhui, China

    Ph.D.
    University of Science and Technology of China
    Statistics
  • 2016 - 2020

    Xi'an, Shaanxi, China

    B.S.
    Northwest University
    Statistics

Awards

  • 2026–2027
    Pyle Fellowship
    Harvard Medical School

    Selected for Fellowship Support to Advance Postdoctoral Research

  • 2025
    National Scholarship for Doctoral Students
    Ministry of Education of China

    Awarded to the Top 3% of Ph.D. Students Nationwide

  • 2020
    Outstanding Graduate Award
    Northwest University

    Awarded to the Top 5% of Graduates

  • 2020
    Annual Best Paper Award
    Journal of Financial Development Research

    Best Paper of the Year

  • 2016–2019
    First-Class Scholarship
    Northwest University

    Highest Scholarship Honor for Undergraduate Students; awarded every academic year.

Professional Service

  • INFORMS Journal on Computing
  • Journal of Business & Economic Statistics
  • Biometrics
  • Statistics and Computing
  • Stat

Research Interests

  • Reinforcement learning and multi-armed bandits
  • Dynamic pricing and assortment optimization
  • Dynamic treatment regimes and heterogeneous effects exploration
  • Large language models (LLMs)

Talks

  • Robust Collaborative Multi-Agent Recommendation With Mixed Sources: A Provably Efficient Bandit Approach. INFORMS Conference on Quality, Statistics, and Reliability (2026 ICQSR), Hong Kong SAR, China, July 8, 2026.
  • Truncated Fusion Learning on Supervised Clustering and Its Fast Stagewise Algorithm. The 3rd Joint Conference on Statistics and Data Science in China (2025 JCSDS), Hangzhou, Zhejiang, China, July 13, 2025.