Support Recovery of Sparse Signals from a Mixture of Linear Measurements

TL;DR

This paper studies the problem of recovering multiple sparse vectors' supports from mixed linear or 1-bit measurements, proposing algorithms that efficiently identify supports with high probability despite measurement noise and mixing.

Contribution

It introduces algorithms for support recovery in mixtures of linear regressions and classifiers, handling noisy, mixed measurements with polynomial and quasi-polynomial measurement complexity.

Findings

Supports of all component vectors can be recovered with high probability.

The algorithms require polynomial in $k, \, ext{log} n$ and quasi-polynomial in $\, ext{log} \, ext{l}$ measurements.

Support recovery is feasible even with measurement noise and random mixing.

Abstract

Recovery of support of a sparse vector from simple measurements is a widely-studied problem, considered under the frameworks of compressed sensing, 1-bit compressed sensing, and more general single index models. We consider generalizations of this problem: mixtures of linear regressions, and mixtures of linear classifiers, where the goal is to recover supports of multiple sparse vectors using only a small number of possibly noisy linear, and 1-bit measurements respectively. The key challenge is that the measurements from different vectors are randomly mixed. Both of these problems have also received attention recently. In mixtures of linear classifiers, the observations correspond to the side of queried hyperplane a random unknown vector lies in, whereas in mixtures of linear regressions we observe the projection of a random unknown vector on the queried hyperplane. The primary step in recovering the unknown vectors from the mixture is to first identify the support of all the individual component vectors. In this work, we study the number of measurements sufficient for recovering the supports of all the component vectors in a mixture in both these models. We provide algorithms that use a number of measurements polynomial in $k, \log n$ and quasi-polynomial in $\ell$, to recover the support of all the $\ell$ unknown vectors in the mixture with high probability when each individual component is a $k$-sparse $n$-dimensional vector.