Yu-Heng Hung (洪鈺恆)

Contact & Profile

I received my Ph.D. in Computer Science from National Yang Ming Chiao Tung University (NYCU) in December 2025, where I was supervised by Professor Ping-Chun Hsieh. My research focuses on reinforcement learning, Bayesian optimization, and bandit learning. Currently, I am a Postdoctoral Fellow (Incoming) at the Georgia Institute of Technology, working with Professor Kai Wang. Before joining NYCU, I received my B.S. in Computer Science from National Tsing Hua University (NTHU) in 2018.

Email: hungyh.cs08@nycu.edu.tw
Mobile: +886953997301
GitHub: github.com/HungYuHeng

Education

  • Ph.D., Computer Science, National Yang Ming Chiao Tung University
    Advisor: Prof. Ping-Chun Hsieh, 2021—2025.
    Dissertation: Learning to Optimize: Bandit Algorithms and Reinforcement Learning for Black-box Decision Problems
  • M.S., Computer Science, National Yang Ming Chiao Tung University
    Advisor: Prof. Ping-Chun Hsieh, 2019—2021.
  • B.S., Computer Science, National Tsing Hua University
    2014—2018.

Professional Experience

  • Visiting Scholar, Georgia Institute of Technology, School of Computational Science and Engineering
    Advisor: Prof. Kai Wang
    Sept 2024 - May 2025, Georgia, United States

Research Interests

  • Reinforcement Learning
  • Bayesian Optimization
  • Bandit Learning
  • Meta Learning

 PUBLICATIONS

Recent Preprints

  • Yu Heng Hung, Kai Wang, Ping-Chun Hsieh, "Non-Stationary Restless Multi-Armed Bandits with Provable Guarantee".
  • Yu-Heng Hung, Ping-Chun Hsieh, Akshay Mete, and P. R. Kumar, "Value-Biased Maximum Likelihood Estimation for Model-based Reinforcement Learning in Discounted Linear MDPs".
  • Yu-Heng Hung and Ping-Chun Hsieh, "Efficient Exploration via Reward-Biased Maximum Likelihood Estimation in Linear Contextual Bandits".

Conference Papers

  • Kuang-Da Wang*, Teng-Ruei Chen*, Yu Heng Hung, Guo-Xun Ko, Shuoyang Ding, Yueh-Hua Wu, Yu-Chiang Frank Wang, Chao-Han Huck Yang, Wen-Chih Peng, and Ping-Chun Hsieh (*: equal contribution), "Test-Time Alignment for Large Language Models via Textual Model Predictive Control", International Conference on Learning Representations (ICLR), 2026 (Acceptance rate = 28%)
  • Yu-Heng Hung, Kai-Jie Lin, Yu-Heng Lin, Chien-Yi Wang, Ping-Chun Hsieh, "BOFormer: Learning to Solve Multi-Objective Bayesian Optimization via Non-Markovian RL", International Conference on Learning Representations (ICLR), 2025 (Acceptance rate = 32%) (An extended version of our ICLR 2024 AutoRL Workshop paper)
  • Yu-Heng Hung, Ping-Chun Hsieh, "Reward-Biased Maximum Likelihood Estimation for Neural Contextual Bandits: A Distributional Learning Perspective", AAAI Conference on Artificial Intelligence (AAAI), 2023 (Acceptance rate = 19.6%)
  • Yu-Heng Hung, Ping-Chun Hsieh, Xi Liu, and P. R. Kumar, "Reward-Biased Maximum Likelihood Estimation for Linear Stochastic Bandits", AAAI Conference on Artificial Intelligence (AAAI), 2021 (Acceptance rate = 21%)
  • Xi Liu, Ping-Chun Hsieh, Yu-Heng Hung, Anirban Bhattacharya, and P. R. Kumar, "Exploration Through Reward Biasing: Reward-Biased Maximum Likelihood Estimation for Stochastic Multi-Armed Bandits", International Conference on Machine Learning (ICML), 2020 (Acceptance rate =22%)

AWARDS

Honors and Awards

  • 2025 Hon Hai Technology Award.
  • Best Poster Honorable Mention at Machine Learning Summer School 2021 Taipei, Aug 2021.
  • Scholarship Program of Ministry of Science and Technology (MOST) to Subsidize Universities and Colleges to Cultivate Outstanding Doctoral Students, Taiwan, Sept 2021.

Work Experience

  • Teaching assistant of Optimization Algorithms in National Yang Ming Chiao Tung University, Sep 2022 - Jan 2023.
  • Teaching assistant of Reinforcement Learning in National Yang Ming Chiao Tung University, Feb 2020 - June 2020.