Oscar Leong
  • NSF Graduate Research Fellow
  • Department of Computational and Applied Mathematics
  • Rice University

I am a fifth year PhD student at Rice University in Computational and Applied Mathematics and a Visiting Scholar at Northeastern University under the supervision of Dr. Paul Hand. I received my undergraduate degree in Mathematics from Swarthmore College. In the summer of 2019, I was an intern at Lawrence Livermore National Laboratory's Data Science Summer Institute working with Dr. Wesam Sakla. I am also a Part-Time Lecturer at Northeastern in the Mathematics department. Lastly, I serve as an Alumni Trustee on the Boston Board of Directors of the non-profit organization Thrive Scholars.

I am interested in solving signal recovery problems with approaches inspired by deep learning and use tools from high dimensional probability, random matrix theory, and optimization to develop provable recovery guarantees for such problems.

News
  • April 2021: I successfully defended my dissertation! Excited to announce I'll be joining Caltech in the Department of Computing + Mathematical Sciences as a von Kármán Instructor in the Fall.

Publications
  • Optimal Sample Complexity of Subgradient Descent for Amplitude Flow via Non-Lipschitz Matrix Concentration (with Paul Hand and Vladislav Voroninski) [arXiv]
    To appear in Communications in Mathematical Sciences
  • Compressive Phase Retrieval: Optimal Sample Complexity with Deep Generative Priors (with Paul Hand and Vladislav Voroninski) [arXiv]
    Submitted
  • Invertible Generative Models for Inverse Problems: Mitigating Representation Error and Dataset Bias (with Muhammad Asim, Max Daniels, Ali Ahmed, and Paul Hand) [arXiv]
    ICML 2020
  • Low Shot Learning with Untrained Neural Networks for Imaging Inverse Problems (with Wesam Sakla) [arXiv]
    NeurIPS 2019 Deep Inverse Workshop. LatinX in AI Oral Presentation.
  • Phase Retrieval Under a Generative Prior (with Paul Hand and Vladislav Voroninski) [arXiv]
    NeurIPS 2018. Oral Presentation (0.6% submissions).
  • Proving Tucker's Lemma With a Volume Argument (with Francis Su, Mutiara Sondjaja, Beauttie Kuture, and Christopher Loa) [arXiv]
    Algebraic and Geometric Methods in Discrete Mathematics, 223-230, Contemp. Math., 685, Amer. Math. Soc., Providences, RI, 2017

Teaching

Northeastern University

  • Instructor of Record (1 session) for MATH 1231: Calculus for Business and Economics (Spring 2020)
  • Instructor of Record (2 sessions) for MATH 1231: Calculus for Business and Economics (Fall 2019)

Rice University

  • Teaching Assistant for CAAM 210: Introduction to Engineering Computation (Fall 2017/Spring 2018)

Richard Tapia Center for Excellence and Equity

Swarthmore College

  • Teaching Assistant for Math 15: Single Variable Calculus (Fall 2013)

Thrive Scholars

Talks

Here is a video of my oral presentation at NeurIPS 2018 on our Deep Phase Retrieval paper along with a link to a 3 minute summary of our work: