Stable safe screening and structured dictionaries for faster ℓ_1 regularization

12/17/2018
by   Cassio Dantas, et al.
8

In this paper, we propose a way to combine two acceleration techniques for the ℓ_1-regularized least squares problem: safe screening tests, which allow to eliminate useless dictionary atoms, and the use of fast structured approximations of the dictionary matrix. To do so, we introduce a new family of screening tests, termed stable screening, which can cope with approximation errors on the dictionary atoms while keeping the safety of the test (i.e. zero risk of rejecting atoms belonging to the solution support). Some of the main existing screening tests are extended to this new framework. The proposed algorithm consists in using a coarser (but faster) approximation of the dictionary at the initial iterations and then switching to better approximations until eventually adopting the original dictionary. A systematic switching criterion based on the duality gap saturation and the screening ratio is derived.Simulation results show significant reductions in both computational complexity and execution times for a wide range of tested scenarios.

READ FULL TEXT
research
12/12/2014

Dynamic Screening: Accelerating First-Order Algorithms for the Lasso and Group-Lasso

Recent computational strategies based on screening tests have been propo...
research
02/08/2021

Dynamic Sasvi: Strong Safe Screening for Norm-Regularized Least Squares

A recently introduced technique for a sparse optimization problem called...
research
05/13/2015

Mind the duality gap: safer rules for the Lasso

Screening rules allow to early discard irrelevant variables from the opt...
research
10/26/2017

Joint Screening Tests for LASSO

This paper focusses on "safe" screening techniques for the LASSO problem...
research
05/19/2014

Screening Tests for Lasso Problems

This paper is a survey of dictionary screening for the lasso problem. Th...
research
09/21/2016

Fast and reliable stereopsis measurement at multiple distances with iPad

Purpose: To present a new fast and reliable application for iPad (ST) fo...
research
03/02/2022

Beyond GAP screening for Lasso by exploiting new dual cutting half-spaces with supplementary material

In this paper, we propose a novel safe screening test for Lasso. Our pro...

Please sign up or login with your details

Forgot password? Click here to reset