Multi-dimensional sparse structured signal approximation using split Bregman iterations

2013/03/21 by Isaac, Yoann, Barthélemy, Quentin, Atif, Jamal +2
#Data Structures and Algorithms (cs.DS) #FOS: Computer and information sciences #Machine Learning (stat.ML)

paper · doi:10.48550/arxiv.1303.5197

Abstract

The paper focuses on the sparse approximation of signals using overcomplete representations, such that it preserves the (prior) structure of multi-dimensional signals. The underlying optimization problem is tackled using a multi-dimensional split Bregman optimization approach. An extensive empirical evaluation shows how the proposed approach compares to the state of the art depending on the signal features.

Related

• Multi-dimensional signal approximation with sparse structured priors\n using split Bregman iterations
• 08492 Executive Summary – Structured Decompositions and Efficient Algorithms
• Minimum Complexity Pursuit: Stability Analysis
• Sparse Approximation by Semidefinite Programming
• Sparse Linear Representation