Real Schur norms and Hadamard matrices

TL;DR

This paper investigates the Schur norms of ±1 matrices, establishing bounds related to Hadamard matrices, and introduces an efficient computational method leading to the discovery of improved almost Hadamard matrices.

Contribution

It provides a bound on Schur norms for ±1 matrices, characterizes when equality occurs, and develops a new computational approach to find better almost Hadamard matrices.

Findings

Schur norm of an n-by-n ±1 matrix is at most √n.

Equality in the bound characterizes Hadamard matrices.

New almost Hadamard matrices outperform previous examples.

Abstract

We present a preliminary study of Schur norms $\|M\|_{S}=\max\{ \|M\circ C\|: \|C\|=1\}$, where M is a matrix whose entries are $\pm1$, and $\circ$ denotes the entrywise (i.e., Schur or Hadamard) product of the matrices. We show that, if such a matrix M is n-by-n, then its Schur norm is bounded by $\sqrt{n}$, and equality holds if and only if it is a Hadamard matrix. We develop a numerically efficient method of computing Schur norms, and as an application of our results we present several almost Hadamard matrices that are better than were previously known.