Dilations, inclusions of matrix convex sets, and completely positive maps
Abstract
A matrix convex set is a set of the form (where each is a set of -tuples of matrices) that is invariant under UCP maps from to and under formation of direct sums. We study the geometry of matrix convex sets and their relationship to completely positive maps and dilation theory. Key ingredients in our approach are polar duality in the sense of Effros and Winkler, matrix ranges in the sense of Arveson, and concrete constructions of scaled commuting normal dilation for tuples of self-adjoint operators, in the sense of Helton, Klep, McCullough and Schweighofer. Given two matrix convex sets and , we find geometric conditions on or on , such that implies that for some constant . For instance, under various symmetry conditions on , we can show that above can be chosen to equal , the number of variables, and in some cases this is sharp. We also find an essentially unique self-dual matrix convex set , the self-dual matrix ball, for which corresponding inclusion and dilation results hold with constant . Our results have immediate implications to spectrahedral inclusion problems studied recently by Helton, Klep, McCullough and Schweighofer. Our constants do not depend on the ranks of the pencils determining the free spectrahedra in question, but rather on the "number of variables" . There are also implications to the problem of existence of (unital) completely positive maps with prescribed values on a set of operators.
Keywords
Cite
@article{arxiv.1601.07993,
title = {Dilations, inclusions of matrix convex sets, and completely positive maps},
author = {Kenneth R. Davidson and Adam Dor-On and Orr Shalit and Baruch Solel},
journal= {arXiv preprint arXiv:1601.07993},
year = {2025}
}
Comments
Proposition 6.3, Corollary 6.4 and Theorem 6.5 in the previous version do not hold in the generality claimed. In this version we added "non-singularity" assumption (Definition 6.3) under which these results hold; in the new version they appear as Proposition 6.7, Corollary 6.8 and Theorem 6.9. 52 pages