Statistical Learning and Inference for Geometric Features on Density and Regression Landscapes
Dr. Wanli Qiao
Statistics Department Seminar
2pm, Wednesday October 20th
CS Conference Room (4201)**, or Zoom (details below)

**If you plan to attend in person, please RSVP here<>**

Geometric features on landscapes, such as ridges, modes, and level sets, are sets that can characterize the structures or shapes of data. Such geometric features are often in the form of low-dimensional manifolds and play an important role in discovering the patterns hidden in point clouds. They have been used as the main concepts in statistical learning tasks such as modal clustering, anomaly detection, and manifold learning. In this talk, I will present novel algorithms of finding geometric features on density and regression landscapes, and their corresponding statistical inference results. The usefulness of geometric feature estimation will be demonstrated in the reconstruction and decomposition of energy landscapes of proteins, as a means of uncovering the organization of structure spaces and discovery of biologically-active macrostates of proteins.

Wanli Qiao is an Assistant Professor in the Department of Statistics at George Mason University. His research areas include modern nonparametric statistics, geometric data analysis, machine learning, and extreme value theory. He has been focusing on tackling theoretical, methodological, and computational challenges arising from statistical learning and inference for geometric features, as well as their applications in molecular biology. He received his PhD degree in Statistics from the University of California, Davis in 2013, and joined George Mason University in 2016 after working in the banking industry for several years. He was a recipient of the Jeffress Trust Award in Interdisciplinary Research in 2017, and his current research is supported by multiple National Science Foundation awards.

Zoom details:
Join Zoom Meeting:

Meeting ID: 989 7748 9545
Passcode: i38sko29
One tap mobile
+13017158592,,98977489545#,,,,*11236197# US (Washington DC)
+12678310333,,98977489545#,,,,*11236197# US (Philadelphia)

Dial by your location
        +1 301 715 8592 US (Washington DC)
        +1 267 831 0333 US (Philadelphia)
Meeting ID: 989 7748 9545
Passcode: 11236197
Find your local number:

Join by SIP
[log in to unmask]<mailto:[log in to unmask]>