Generalized Nullspace Property for Structurally Sparse Signals

10/12/2019
by   Mustafa D. Kaba, et al.
0

We propose a new framework for studying the exact recovery of signals with structured sparsity patterns, which generalizes the well-known nullspace property for standard sparse recovery. We show that for each dictionary there is a maximum sparsity pattern—described by a mathematical object called an "abstract simplicial complex"—that can be recovered via ℓ_1-minimization. We provide two different characterizations of this maximum sparsity pattern, one based on extreme points and the other based on vectors of minimal support. In addition, we show how this new framework is useful in the study of sparse recovery problems when the dictionary takes the form of a graph incidence matrix or a partial discrete Fourier transform. In both cases we successfully characterize the collection of all support sets for which exact recovery via ℓ_1-minimization is possible. As a by product, we show that when the dictionary is an incidence matrix, exact recovery can be certified in polynomial time, although this is known to be NP-hard for general matrices.

READ FULL TEXT
POST COMMENT

Comments

There are no comments yet.

Authors

page 8

03/26/2018

Sparse Recovery over Graph Incidence Matrices: Polynomial Time Guarantees and Location Dependent Performance

Classical results in sparse recovery guarantee the exact reconstruction ...
03/04/2019

Piecewise Sparse Recovery in Unions of Bases

Sparse recovery is widely applied in many fields, since many signals or ...
10/04/2021

Identifiability in Exact Two-Layer Sparse Matrix Factorization

Sparse matrix factorization is the problem of approximating a matrix Z b...
10/19/2020

Sparse Recovery Analysis of Generalized J-Minimization with Results for Sparsity Promoting Functions with Monotonic Elasticity

In this paper we theoretically study exact recovery of sparse vectors fr...
05/20/2014

Sparsity Based Methods for Overparameterized Variational Problems

Two complementary approaches have been extensively used in signal and im...
09/24/2018

Representing Sets as Summed Semantic Vectors

Representing meaning in the form of high dimensional vectors is a common...
11/14/2019

Predicting sparse circle maps from their dynamics

The problem of identifying a dynamical system from its dynamics is of gr...
This week in AI

Get the week's most popular data science and artificial intelligence research sent straight to your inbox every Saturday.