Office:06-114

Department of Statistics and Applied Probability
National University of Singapore
21 Lower Kent Ridge Road             
Singapore 117546
Email:
zhigang.yao@nus.edu.sg
[+]
65 6601-3125                                

02

 

Few words about me:


Zhigang Yao
is an assistant professor in the Department of Statistics and Applied Probability at the National University of Singapore (NUS). He received his Ph.D. in Statistics from University of Pittsburgh in 2011. His thesis advisors are Bill Eddy at Carnegie Mellon and Leon Gleser at University of Pittsburgh. Before joining NUS, he has been working with Victor Panaretos as a post-doc researcher at the Swiss Federal Institute of Technology (EPFL) from 2011-2014.

 

 

Few words about my work:

My main research area is statistical inference for complex data. Broadly speaking, in terms of statistical complexity, there are two categories of "Complex Data" that I have been interested in. The first one arises because of temporal/spatial structure in the collection of the data, often with added complexity because data are incomplete, in terms of variables which may be missing, or poorly measured. The second one has emerged as the data is gathered with increasingly more dimensions while there are only tens or hundreds of instances available for study or the data itself lies in non-Euclidean space.

My main work in the first category includes inverse problem from brain imaging (i.e., MEG) and tomographic reconstruction (i.e., electron microscope). In MEG, the problem is to localize the electrical source in the brain using the extremely weak magnetic signal outside of the head; in tomography, the problem is to obtain the complete 3D folding of the particle from the partial knowledge (say, its 2D projections) recorded on the film.

My work in the second category includes making statistical inference by exploiting useful structure (i.e., sparse features) in high-dimensional data where the useful signal is rare and weak, and by finding principal variation (principal flows or sub-manifolds) of data lying on manifolds. For either of such data, most conventional approaches fail.