English

Distance matrices of subsets of the Hamming cube

Functional Analysis 2020-08-03 v2

Abstract

Graham and Winkler derived a formula for the determinant of the distance matrix of a full-dimensional set of n+1n + 1 points {x0,x1,,xn}\{ x_{0}, x_{1}, \ldots , x_{n} \} in the Hamming cube Hn=({0,1}n,1)H_{n} = ( \{ 0,1 \}^{n}, \ell_{1} ). In this article we derive a formula for the determinant of the distance matrix DD of an arbitrary set of m+1m + 1 points {x0,x1,,xm}\{ x_{0}, x_{1}, \ldots , x_{m} \} in HnH_{n}. It follows from this more general formula that det(D)0\det (D) \not= 0 if and only if the vectors x0,x1,,xmx_{0}, x_{1}, \ldots , x_{m} are affinely independent. Specializing to the case m=nm = n provides new insights into the original formula of Graham and Winkler. A significant difference that arises between the cases m<nm < n and m=nm = n is noted. We also show that if DD is the distance matrix of an unweighted tree on n+1n + 1 vertices, then D11,1=2/n\langle D^{-1} \mathbf{1}, \mathbf{1} \rangle = 2/n where 1\mathbf{1} is the column vector all of whose coordinates are 11. Finally, we derive a new proof of Murugan's classification of the subsets of HnH_{n} that have strict 11-negative type.

Keywords

Cite

@article{arxiv.2005.09205,
  title  = {Distance matrices of subsets of the Hamming cube},
  author = {Ian Doust and Gavin Robertson and Alan Stoneham and Anthony Weston},
  journal= {arXiv preprint arXiv:2005.09205},
  year   = {2020}
}

Comments

11 pages