QuaCaeLimSch10.pdf (2.4 MB)
Convex relaxation of mixture regression with efficient algorithms
conference contribution
posted on 2023-06-08, 16:44 authored by Novi QuadriantoNovi Quadrianto, Tiberio S Caetano, John Lim, Dale SchuurmansWe 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.
History
Publication status
- Published
File Version
- Accepted version
Journal
Proceedings of the 23rd annual conference on Neural Information Processing Systems 2009; Vancouver, Vancouver, Canada; 6 - 11 December 2009Publisher
CurranPublisher URL
Issue
22Volume
1Page range
1491-1499Pages
2348.0Book title
Neural information processing systems: 23rd annual conference on neural information processing systems 2009Place of publication
Red Hook, NYISBN
9781615679119Series
Advances in neural information processing systemsDepartment affiliated with
- Informatics Publications
Full text available
- Yes
Peer reviewed?
- Yes