Pengkun Yang
Assistant Professor
Research Areas: theory and algorithms for high-dimensional statistics, mathematical data science, statistical machine learning, optimization
Office: Room 209-D, Weiqing Building, Tsinghua University
Phone: +86-10-62787138
Email: yangpengkun@tsinghua.edu.cn
Work Experiences:
- 2021-03 - now: Assistant Professor at the Center for Statistical Science, Tsinghua University
- 2018-10 - 2020-10: Postdoc at the Department of Electrical Engineering, Princeton University
Education:
- 2016-05 - 2018-08: Ph.D. in ECE, UIUC
- 2013-08 - 2016-05: M.S. in ECE, UIUC
- 2009-08 - 2013-07: B.E. in EE, Tsinghua University
Monographs:
- Polynomial methods in statistical inference: theory and practice, with Yihong Wu, Foundations and Trends® in Communications and Information Theory, now publisher, 2020.
Selected publications:
- Optimal estimation of Gaussian mixtures via denoised method of moments. with Yihong Wu, Annals of Statistics, Vol. 48, No. 4, pp. 1981-2007, 2020. [code]
- On Learning Over-parameterized Neural Networks: A Functional Approximation Perspective, with Lili Su, NeurIPS 2019.
- Chebyshev polynomials, moment matching, and optimal estimation of the unseen, with Yihong Wu, Annals of Statistics, Vol. 47, No. 2, pp. 857-883, 2019. [code]
- Minimax Rates of Entropy Estimation on Large Alphabets via Best Polynomial Approximation, with Yihong Wu, IEEE Transactions on Information Theory, Vol. 62, Issue, 6, pp. 3702-3720, 2016. [code]
- Optimal entropy estimation on large alphabets via best polynomial approximation, with Yihong Wu, ISIT 2015. Best Student Paper Award
Awards: