Weijie Su


Hi, I am an Assistant Professor in the Wharton Statistics Department and, by courtesy, in the Department of Computer and Information Science, at the University of Pennsylvania. I serve as a co-director of Penn Research in Machine Learning. Prior to joining Penn in Summer 2016, I obtained my Ph.D. from Stanford University in 2016, under the supervision of Emmanuel Cand├Ęs. I received my bachelor's degrees from Peking University in 2011.

My Google Scholar profile and CV.


Office: 472 Jon M. Huntsman Hall
3730 Walnut Street
Philadelphia, PA 19104

Email: suw AT wharton DOT upenn DOT edu

Recent news

  • A new paper that proposes the Layer-Peeled Model toward understanding deep learning. Unlike most models (such as two-layer neural networks) that are only capable of explaining already observed phenomena of deep learning models, our Layer-Peeled Model can predict a new surprise of deep learning!

  • Two new papers that use local elasticity for a phenomenological understanding of the effectiveness of deep learning: paper 1 (NeurIPS 2020) and paper 2 (ICML 2021).

  • Privacy analysis of deep learning using stochastic gradient descent in the GDP framework is now available in TensorFlow. Check out the code and the tutorial for the paper. The maintainer is Zhiqi Bu.