English

Tensor network ranks

Numerical Analysis 2019-02-12 v2

Abstract

In problems involving approximation, completion, denoising, dimension reduction, estimation, interpolation, modeling, order reduction, regression, etc, we argue that the near-universal practice of assuming that a function, matrix, or tensor (which we will see are all the same object in this context) has \emph{low rank} may be ill-justified. There are many natural instances where the object in question has high rank with respect to the classical notions of rank: matrix rank, tensor rank, multilinear rank --- the latter two being the most straightforward generalizations of the former. To remedy this, we show that one may vastly expand these classical notions of ranks: Given any undirected graph GG, there is a notion of GG-rank associated with GG, which provides us with as many different kinds of ranks as there are undirected graphs. In particular, the popular tensor network states in physics (e.g., \textsc{mps}, \textsc{ttns}, \textsc{peps}) may be regarded as functions of a specific GG-rank for various choices of GG. Among other things, we will see that a function, matrix, or tensor may have very high matrix, tensor, or multilinear rank and yet very low GG-rank for some GG. In fact the difference is in the orders of magnitudes and the gaps between GG-ranks and these classical ranks are arbitrarily large for some important objects in computer science, mathematics, and physics. Furthermore, we show that there is a GG such that almost every tensor has GG-rank exponentially lower than its rank or the dimension of its ambient space.

Keywords

Cite

@article{arxiv.1801.02662,
  title  = {Tensor network ranks},
  author = {Ke Ye and Lek-Heng Lim},
  journal= {arXiv preprint arXiv:1801.02662},
  year   = {2019}
}

Comments

37 pages, 7 figures