Block based refitting in $\ell_{12}$ sparse regularisation

TL;DR

This paper introduces a novel refitting framework for $ ext{l}_{12}$ block regularization in linear regression, effectively reducing bias while maintaining sparsity support, with applications demonstrated in image restoration models.

Contribution

It proposes a new refitting block penalty and an algorithm to improve $ ext{l}_{12}$ regularized solutions, addressing bias issues in sparse recovery tasks.

Findings

The new penalty effectively reduces bias in $ ext{l}_{12}$ solutions.

The algorithm successfully refits solutions in TV and TGV models.

Experiments show improved support preservation and solution accuracy.

Abstract

In many linear regression problems, including ill-posed inverse problems in image restoration, the data exhibit some sparse structures that can be used to regularize the inversion. To this end, a classical path is to use $\ell_{12}$ block based regularization. While efficient at retrieving the inherent sparsity patterns of the data -- the support -- the estimated solutions are known to suffer from a systematical bias. We propose a general framework for removing this artifact by refitting the solution towards the data while preserving key features of its structure such as the support. This is done through the use of refitting block penalties that only act on the support of the estimated solution. Based on an analysis of related works in the literature, we introduce a new penalty that is well suited for refitting purposes. We also present a new algorithm to obtain the refitted solution along with the original (biased) solution for any convex refitting block penalty. Experiments illustrate the good behavior of the proposed block penalty for refitting solutions of Total Variation and Total GeneralizedVariation models.