Statistics – Machine Learning
Scientific paper
2010-05-26
Statistics
Machine Learning
27 pages. Previous Version Name: An Efficient Proximal-Gradient Method for Single and Multi-task Regression with Structured Sp
Scientific paper
We study the problem of estimating high dimensional regression models regularized by a structured-sparsity-inducing penalty that encodes prior structural information on either input or output sides. We consider two widely adopted types of such penalties as our motivating examples: 1) overlapping-group-lasso penalty, based on the $\ell_1/\ell_2$ mixed-norm penalty, and 2) graph-guided fusion penalty. For both types of penalties, due to their non-separability, developing an efficient optimization method has remained a challenging problem. In this paper, we propose a general optimization approach, called smoothing proximal gradient method, which can solve the structured sparse regression problems with a smooth convex loss and a wide spectrum of structured-sparsity-inducing penalties. Our approach combines the smoothing technique and the proximal gradient method. It achieves a convergence rate significantly faster than the standard first-order method, subgradient method, and is much more scalable than the most widely used interior-point method. The efficiency and scalability of our method are demonstrated on both simulated and real genetic datasets.
Carbonell Jaime G.
Chen Xi
Kim Seyoung
Lin Qihang
Xing Eric P.
No associations
LandOfFree
An Efficient Proximal Gradient Method for General Structured Sparse Learning does not yet have a rating. At this time, there are no reviews or comments for this scientific paper.
If you have personal experience with An Efficient Proximal Gradient Method for General Structured Sparse Learning, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and An Efficient Proximal Gradient Method for General Structured Sparse Learning will most certainly appreciate the feedback.
Profile ID: LFWR-SCP-O-604701