An elementary approach to the problem of column selection in a rectangular matrix

TL;DR

This paper presents a simple, constructive method for selecting well-conditioned submatrices from a given rectangular matrix with normalized columns, improving bounds on singular values with minimal logarithmic loss.

Contribution

It introduces an elementary, fully constructive approach to extract well-conditioned submatrices, providing individual bounds for each singular value with only a logarithmic factor loss.

Findings

Provides explicit bounds for each singular value of the submatrix.

Method is fully constructive and elementary in nature.

Achieves near-optimal bounds with minimal logarithmic loss.

Abstract

The problem of extracting a well conditioned submatrix from any rectangular matrix (with normalized columns) has been studied for some time in functional and harmonic analysis; see \cite{BourgainTzafriri:IJM87,Tropp:StudiaMath08,Vershynin:IJM01} for methods using random column selection. More constructive approaches have been proposed recently; see the recent contributions of \cite{SpielmanSrivastava:IJM12,Youssef:IMRN14}. The column selection problem we consider in this paper is concerned with extracting a well conditioned submatrix, i.e. a matrix whose singular values all lie in $[1-\epsilon,1+\epsilon]$. We provide individual lower and upper bounds for each singular value of the extracted matrix at the price of conceding only one log factor in the number of columns, when compared to the Restricted Invertibility Theorem of Bourgain and Tzafriri. Our method is fully constructive and the proof is short and elementary.