Convex relaxation of mixture regression with efficient algorithms

Quadrianto, Novi, Caetano, Tiberio S, Lim, John and Schuurmans, Dale (2009) Convex relaxation of mixture regression with efficient algorithms. Published in: Bengio, Yoshua, Schuurmans, Dale, Lafferty, John, Williams, Chris and Culotta, Aron, (eds.) Proceedings of the 23rd annual conference on Neural Information Processing Systems 2009; Vancouver, Vancouver, Canada; 6 - 11 December 2009. 1 (22) 1491-1499. Curran, Red Hook, NY. ISBN 9781615679119

PDF - Accepted Version
Download (2MB) | Preview


We develop a convex relaxation of maximum a posteriori estimation of a mixture of regression models. Although our relaxation involves a semidefinite matrix variable, we reformulate the problem to eliminate the need for general semidefinite programming. In particular, we provide two reformulations that admit fast algorithms. The first is a max-min spectral reformulation exploiting quasi-Newton descent. The second is a min-min reformulation consisting of fast alternating steps of closed-form updates. We evaluate the methods against Expectation-Maximization in a real problem of motion segmentation from video data.

Item Type: Conference Proceedings
Schools and Departments: School of Engineering and Informatics > Informatics
Subjects: Q Science > Q Science (General)
Depositing User: Novi Quadrianto
Date Deposited: 24 Feb 2014 11:21
Last Modified: 19 Jun 2017 11:47

View download statistics for this item

📧 Request an update