Coordinate Descent Algorithms (Joint CS/Math Colloquium)
Speaker: Stephen J. Wright, University of Wisconsin
Location: Warren Weaver Hall 1302
Date: December 7, 2015, 3:45 p.m.
Host: Michael Overton
There is renewed interest in elementary methods for optimization, sparked in large part by applications in data analysis. Prominent among these methods is coordinate descent, in which functions are minimized by searching successively along coordinate directions or hyperplanes. Several practical extensions of this approach have been proposed, including efficient implementations of Nesterov acceleration, randomized versions, asynchronous parallel execution, and extensions to nonsmooth objectives. New theory has improved our understanding of the convergence behavior of these methods, though numerous aspects remain unexplained. This talk will survey these recent developments, outline problem classes in which the approach is useful, and describe several open questions.
In-person attendance only available to those with active NYU ID cards. All individuals must show the Daily Screener green pass in order to gain entry to the building.