Safe squeezing for antisparse coding

11/18/2019
by   Clément Elvira, et al.
0

Spreading the information over all coefficients of a representation is a desirable property in many applications such as digital communication or machine learning. This so-called antisparse representation can be obtained by solving a convex program involving an ℓ_∞-norm penalty combined with a quadratic discrepancy. In this paper, we propose a new methodology, dubbed safe squeezing, to accelerate the computation of antisparse representation. We describe a test that allows to detect saturated entries in the solution of the optimization problem. The contribution of these entries is compacted into a single vector, thus operating a form of dimensionality reduction. We propose two algorithms to solve the resulting lower dimensional problem. Numerical experiments show the effectiveness of the proposed method to detect the saturated components of the solution and illustrates the induced computational gains in the resolution of the antisparse problem.

READ FULL TEXT

page 1

page 2

page 3

page 4

research
06/04/2023

Prescriptive PCA: Dimensionality Reduction for Two-stage Stochastic Optimization

In this paper, we consider the alignment between an upstream dimensional...
research
10/22/2021

Safe rules for the identification of zeros in the solutions of the SLOPE problem

In this paper we propose a methodology to accelerate the resolution of t...
research
02/28/2023

Safe peeling for l0-regularized least-squares with supplementary material

We introduce a new methodology dubbed “safe peeling” to accelerate the r...
research
02/17/2020

Shifted Lanczos method for quadratic forms with Hermitian matrix resolvents

Quadratic forms of Hermitian matrix resolvents involve the solutions of ...
research
06/16/2021

Non-PSD Matrix Sketching with Applications to Regression and Optimization

A variety of dimensionality reduction techniques have been applied for c...
research
01/03/2018

Clustering of Data with Missing Entries

The analysis of large datasets is often complicated by the presence of m...
research
01/06/2019

Solving large-scale L1-regularized SVMs and cousins: the surprising effectiveness of column and constraint generation

The linear Support Vector Machine (SVM) is one of the most popular binar...

Please sign up or login with your details

Forgot password? Click here to reset