Yicheng Li

prof_pic.jpg

李易诚 | Yicheng Li

Zijiang Young Scholar

School of Statistics, ECNU

ycli@sfs.ecnu.edu.cn

Google Scholar

I am Yicheng Li (李易诚), a Zijiang Young Scholar (Top Young Talent) in the School of Statistics at East China Normal University. I received my Ph.D. in Statistics from the Department of Statistics and Data Science at Tsinghua University, advised by Professor Qian Lin, and my B.Sc. in Pure and Applied Mathematics from the Department of Mathematical Sciences at Tsinghua University.

My main research areas are deep learning theory and statistical learning theory, with a particular focus on feature learning, neural tangent kernel theory, kernel methods, differential privacy, and related topics. See my Google Scholar page.

If you are interested in my research, feel free to contact me by email.

Academic Appointments

  • 2026.7–Present: Zijiang Young Scholar (Top Young Talent), School of Statistics, East China Normal University.

Education

  • 2022–2026.6: Ph.D. in Statistics, Department of Statistics and Data Science, Tsinghua University. Advisor: Qian Lin. Thesis (Tsinghua University Outstanding Doctoral Dissertation): Understanding the Generalization of Neural Networks: From Neural Tangent Kernel Regression to Adaptive Feature Models.
  • 2025.6–2025.12: Visiting Student, Department of Statistics, The Wharton School, University of Pennsylvania. Host: Tony Cai.
  • 2018–2022: B.Sc. in Pure and Applied Mathematics, Department of Mathematical Sciences, Tsinghua University; minor in Data Science; Tsinghua University Outstanding Graduate.

Selected publications

  1. Minimax and Adaptive Covariance Matrix Estimation under Differential Privacy
    T. Tony Cai and Yicheng Li (alphabetical order)
    arXiv preprint, arXiv:2603.19703, Mar 2026
    arXiv
  2. Several Supporting Evidences for the Adaptive Feature Program
    Yicheng Li and Qian Lin
    arXiv preprint, arXiv:2511.09425, Nov 2025
    arXiv
  3. Diagonal Over-Parameterization in Reproducing Kernel Hilbert Spaces as an Adaptive Feature Model: Generalization and Adaptivity
    Yicheng Li and Qian Lin
    arXiv preprint, arXiv:2501.08679, Jan 2025
    arXiv
  4. Generalization Error Curves for Analytic Spectral Algorithms under Power-Law Decay
    Yicheng Li, Weiye Gan, Zuoqiang Shi, and Qian Lin
    arXiv preprint, arXiv:2401.01599, Jan 2024
    arXiv
  5. On the Saturation Effect of Kernel Ridge Regression
    Yicheng Li, Haobo Zhang, and Qian Lin
    In 11th International Conference on Learning Representations (ICLR 2023), Feb 2023
    arXiv OpenReview

Miscellaneous

  • President (2022–2023) and Vice President (2023–2025) of the Tsinghua University Student Reading Club.
  • Contributor to TeXiFy-IDEA, an open-source LaTeX plugin for the IntelliJ Platform.