May 04, 2012
12:30 PM - 2:00 PM

IEOR-DRO Seminar: Prof. Elad Hazan (Technion, Israel). Sublinear Optimization for Machine Learning.

Directions and Map

Print this event

Add event to my calendar

Title: Sublinear Optimization for Machine Learning
Speaker: Elad Hazan (Technion, Israel)
Abstract: Linear classification is a fundamental problem of machine learning, in which positive and negative examples of a concept are represented in Euclidean space by their feature vectors, and we seek to find a hyperplane separating the two classes of vectors. We'll present the first sublinear-time algorithms for linear classification, support vector machine training, and other related optimization problems, including general semi-definite programming. These new algorithms are based on a primal-dual approach, and use a combination of novel sampling techniques and the randomized implementation of online learning algorithms. We give lower bounds which show our running times to be nearly best possible in the unit-cost RAM model.
Registration is not required

For more information please e-mail Karin Jihk-Engelcrona.