*********************************
There is now a CONTENT FREEZE for Mercury while we switch to a new platform. It began on Friday, March 10 at 6pm and will end on Wednesday, March 15 at noon. No new content can be created during this time, but all material in the system as of the beginning of the freeze will be migrated to the new platform, including users and groups. Functionally the new site is identical to the old one. webteam@gatech.edu
*********************************
Atlanta, GA | Posted: November 3, 2009
Charlie Brubaker, an August 2009 Ph.D. graduate in Computer Science, has won the College of Computing Doctoral Dissertation Award for his thesis, “Extensions of Principal Component Analysis,” and the College has voted to recommend the work for the ACM Doctoral Dissertation Award, which each year recognizes the best CS dissertation in the country.
Brubaker was advised by Distinguished Professor Santosh Vempala (CS), and his thesis committee includes professors Haesun Park (CSE), Vladimir Koltchinskii (Math), Adam Kalai, senior researcher at Microsoft Research New England, and Ravi Kannan, principal researcher for Microsoft Research India.
“My journey through the Ph.D. program has been a long one, taking me from robotics, to computer vision, to machine learning, and finally to theory,” Brubaker said. “Making so many transitions slowed my graduation, no doubt, but it also allowed me to work with leaders across several research fields and has given me a better perspective on the research world. Ultimately, I am the better for it, and I am grateful to everyone I have had the opportunity to work with.”
Principal Component Analysis (PCA) is one of the most widely used data-analysis techniques in science and engineering. In a letter explaining the significance of Brubaker’s research, Vempala said the thesis (available here) made three distinct contributions to the field: it gives a new notion of principal components called “Isotropic PCA” that is scale-invariant; it suggests a new method called “Robust PCA” for dealing with outliers and noisy data; and it extends the notion of principal components from matrices to tensors (multi-dimensional arrays). PCA is notorious for its vulnerability to noise and for its lack of scale-invariance. The thesis addresses these outstanding issues and will make PCA more practical and reliable.
“Besides the sheer technical strength of his results, Charlie has put in a great deal of effort into clear exposition, and has included multiple chapters with introductory and background material,” Vempala wrote. “His thesis is consistently insightful and a pleasure to read. It is hard for me to recall a more impressive thesis in algorithms in recent years.”
“What strikes me as most impressive about Charlie’s work is that it so beautifully has a foot in absolutely first-rate theory, as well as practical, broad and important application areas,” CS Chair and Professor Ellen Zegura wrote in her letter supporting Brubaker’s nomination for the ACM award. “In the next several years, I think we will see numerous uses of and citations to Charlie’s work in many domains.”
Brubaker and his wife, Stephanie, have two children, Charles and Sophie.
