Robust Modeling and Analysis of High-Dimensional Data
Speaker: Dr. John Wright, Microsoft Research
Series: Electrical Engineering Departmental Seminar
Location: Engineering Quadrangle B205
Date/Time: Monday, February 14, 2011, 4:30 p.m. - 5:30 p.m.
In this talk, I introduce several recent theoretical and algorithmic advances in robust recovery of low-dimensional structure from high-dimensional data. I show how to correctly and efficiently recover two important, closely-related types of low-dimensional structure: sparse vectors and low-rank matrices. For sparse vectors, we prove that as long as the signal of interest has a sufficiently sparse representation in a coherent dictionary, convex programming corrects large fractions of errors. In the same spirit, we prove that convex programming recovers low-rank matrices from large fractions of errors and missing observations. I motivate these general problems from the perspective of automatic face recognition in computer vision, and demonstrate how theoretical advances have inspired progress on this challenging problem. I discuss several additional applications of these tools including robust batch image alignment and registration, 3D shape recovery from multiple images, video stabilization and enhancement, web data analysis, indexing and search.
John Wright received his PhD in Electrical Engineering from the University of Illinois at Urbana-Champaign in October 2009. He is currently a researcher in the Visual Computing group at Microsoft Research Asia. His research focuses on developing provably correct and efficient tools for recovering low-dimensional structure in high-dimensional datasets, even when data are missing or grossly corrupted. These techniques address critical estimation problems in imaging and vision applications such as automatic face recognition, video stabilization and tracking, image and data segmentation, and more. They also find application outside of vision, for example in web data analysis and bioinformatics. His work has received a number of awards and honors, including the 2009 Lemelson-Illinois Prize for Innovation for his work on robust face recognition, the 2009 UIUC Martin Award for Excellence in Graduate Research, a 2008-2010 Microsoft Research Fellowship, a Carver fellowship, and a UIUC Bronze Tablet award.