Poster
Convex Relaxation of Mixture Regression with Efficient Algorithms
Novi Quadrianto · Tiberio Caetano · John Lim · Dale Schuurmans

Wed Dec 9th 07:00 -- 11:59 PM @ None #None

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.

Author Information

Novi Quadrianto (University of Sussex and HSE)
Tiberio Caetano (NICTA Canberra)
John Lim
Dale Schuurmans (Google Brain & University of Alberta)

More from the Same Authors