My research aims to develop a new generation of inference methods and theory for modern statistics and machine learning, especially focusing on:
- Combinatorial functionals like connectivity, degree, and other topological structures of graphs, ranking, clustering, hyper graphs, etc;
- Complex data structures like high dimensionality, heterogeneity, nonlinearity, heavy-tailness, time-dependency, etc;
- Complicated algorithms like distributed algorithms, nonconvex optimization, kernel methods, etc.
With the problems above, I am interested in studying the uncertainty assessment methodology, probabilistic universality phenomenon, and information-theoretical lower bound theory. My research finds main applications in computational neuroscience.
Ph.D., 2018, Princeton University