Office: IPAM 1129, UCLA.
Email: fei dot feng at math dot ucla dot edu

About Me

I am a fourth-year Ph.D. candidate at UCLA Math Department. My advisor is Dr. Wotao Yin. My research interests are optimization, reinforcement learning (RL), and parallel computing. I enjoy applying mathematics theory to solve realistic problems.

Currently, I am working on

  1. Transfer RL
  2. Function approximation in RL
  3. Asynchronous parallel RL

I have a lot of hobbies, including singing, dancing, traveling, photography, painting, and reading. I love my life and appreciate every little beauty in this world.

Updated on Dec. 5, 2019.

Publications and Presentations

  1. Provably Efficient Exploration for RL with Unsupervised Learning.[Preprint]

    Fei Feng, Ruosong Wang, Wotao Yin, Simon S. Du, Lin F. Yang

  2. Does Knowledge Transfer Always Help to Learn A Better Policy? [Preprint]

    Fei Feng, Wotao Yin, Lin F. Yang

  3. AsyncQVI: Async-parallel Q-Value Iteration for RL. [AISTATS 2020][Code]

    Yibo Zeng, Fei Feng, Wotao Yin

    • INFORMS 2019 Seattle [Talk Slides]
    • 2019 Southern California Applied Mathematics Symposium (SOCAMS) [Poster]
  4. Acceleration of SVRG and Katyusha X by Inexact Preconditioning. [ICML 2019][Code]

    Yanli Liu, Fei Feng, Wotao Yin

  5. A2BCD: Asynchronous Acceleration with Optimal Complexity. [top-rated at ICLR 2019]

    Robert Hannah, Fei Feng, Wotao Yin


  1. Very good: C++/C, JAVA, MPI, MATLAB
  2. Moderate: Python, Javascript, HTML, SQL
  3. Libraries: BLAS/LAPACK, EIGEN, GSL
  4. Languages: English, Chinese, Cantonese, Basic German