Title: Graph-based Change-point Detection for Non-Euclidean and Multivariate Data
We present a new framework for the detection of change-points. The approach utilizes similarity information among observations and can be applied to sequences of multivariate or non-Euclidean data. Both offline and online detection will be discussed and analytical ways to determine the threshold of claiming a change are also provided, making the methods easy-off-the-shelf tools for real applications.
Lynna Chu is currently an Assistant Professor in the Department of Statistics at Iowa State University. She completed her PhD in Biostatistics from UC Davis under the supervision of Hao Chen. Before that, she completed her undergraduate at UCLA.