Low-dimensional Modeling for Deep Learning, IEEE Denver Computer, Information Theory, and Robotics Society, June, 2022.
Geometric Analysis of Neural Collapse with Unconstrained Features, The 2nd Workshop on Seeking Low-dimensionality in Deep Neural Networks (SLowDNN), November, 2021.
Robust Recovery with Over-parameterized Model, Microsoft Research Day, Microsoft, March, 2021.
Provable Nonsmooth Nonconvex Approaches for Low-Dimensional Models, Statistics, Optimization and Machine Learning Seminar, University of Colorado Boulder, January, 2020.
Provable Nonconvex Approaches for Low-rank Models, invited speaker at Workshop on Low-Rank Models and Applications (LRMA), University of Mons, Belgium, Sep 12 – 13, 2019.