Sparse Optimization on General Atomic Sets: Greedy and Forward-Backward Algorithms

TL;DR

This paper extends greedy and forward-backward algorithms to optimize over general atomic sets, providing strong approximation guarantees and introducing a new concept called the sparse atomic condition number.

Contribution

It generalizes sparse atomic optimization to broad atomic sets, establishing linear convergence rates and approximation guarantees for greedy and forward-backward algorithms.

Findings

Greedy algorithms achieve strong approximation guarantees on general atomic sets.

Introduction of the sparse atomic condition number for convergence analysis.

Forward-backward algorithms match greedy algorithms in approximation guarantees.

Abstract

We consider the problem of sparse atomic optimization, where the notion of "sparsity" is generalized to meaning some linear combination of few atoms. The definition of atomic set is very broad; popular examples include the standard basis, low-rank matrices, overcomplete dictionaries, permutation matrices, orthogonal matrices, etc. The model of sparse atomic optimization therefore includes problems coming from many fields, including statistics, signal processing, machine learning, computer vision and so on. Specifically, we consider the problem of maximizing a restricted strongly convex (or concave), smooth function restricted to a sparse linear combination of atoms. We extend recent work that establish linear convergence rates of greedy algorithms on restricted strongly concave, smooth functions on sparse vectors to the realm of general atomic sets, where the convergence rate involves a novel quantity: the "sparse atomic condition number". This leads to the strongest known multiplicative approximation guarantees for various flavors of greedy algorithms for sparse atomic optimization; in particular, we show that in many settings of interest the greedy algorithm can attain strong approximation guarantees while maintaining sparsity. Furthermore, we introduce a scheme for forward-backward algorithms that achieves the same approximation guarantees. Secondly, we define an alternate notion of weak submodularity, which we show is tightly related to the more familiar version that has been used to prove earlier linear convergence rates. We prove analogous multiplicative approximation guarantees using this alternate weak submodularity, and establish its distinct identity and applications.