About Me

I am currently a post-doctoral research associate in the department of Data Science of Operations at the Marshall School of Business, University of Southern California (USC), advised by Professor Vishal Gupta. I obtained my Ph.D. degree from the  Daniel J. Epstein Department of Industrial and Systems Engineering, where I was advised by Professor Jong-Shi Pang.

My research focuses on advancing the fundamental optimization theory and designing fast, scalable algorithms that can be suitably deployed to a wide spectrum of applications in machine learning, data science, operations research, econometrics and decision science. With an emphasis on nonconvex modelings under uncertainty, the goal of my research is to build a rigorous mathematical foundation for more robust and generalizable analytics, so that essential problems that lie in the center of progressing our society, industry and business can be solved faithfully and efficiently. Under this driving goal, I am interested in the following topics:

1. Nonconvex and nonsmooth optimization in statistical learning with provable guarantees to the sharp computable solutions,

   • deep learning,  sparsity learning,  hyperparameter selections.

2. Efficient combinations of sampling and surrogation methods for nonconvex stochastic programs,

   • resource allocation, Bayesian inference, portfolio selection.

3. Robust optimization for nonconvex statistical and economic models against noise and adversarial attacks,

   • adversarial deep learning, robust complimentarity problems.

The applications that I am interested in: machine learning, policy learning, power systems and public health.

Interests: nonconvex optimization under uncertainty for data science and operations research.