Tensorization of quasi-Hilbertian Sobolev spaces

TL;DR

This paper resolves the tensorization problem for Sobolev spaces on product metric measure spaces in the case p=2, showing the equivalence of two natural definitions when spaces are infinitesimally quasi-Hilbertian.

Contribution

It proves the equality of Sobolev spaces on product spaces with two definitions for p=2 in the quasi-Hilbertian setting, extending understanding of Sobolev space tensorization.

Findings

W^{1,2}(X×Y) = J^{1,2}(X,Y) for quasi-Hilbertian spaces

Norms agree on algebraic tensor product for p in (1,∞)

Density of tensor products in Sobolev spaces established for p=2

Abstract

The tensorization problem for Sobolev spaces asks for a characterization of how the Sobolev space on a product metric measure space $X\times Y$ can be determined from its factors. We show that two natural descriptions of the Sobolev space from the literature coincide, $W^{1,2}(X\times Y)=J^{1,2}(X,Y)$, thus settling the tensorization problem for Sobolev spaces in the case $p=2$, when $X$ and $Y$ are infinitesimally quasi-Hilbertian, i.e. the Sobolev space $W^{1,2}$ admits an equivalent renorming by a Dirichlet form. This class includes in particular metric measure spaces $X,Y$ of finite Hausdorff dimension as well as infinitesimally Hilbertian spaces. More generally for $p\in (1,\infty)$ we obtain the norm-one inclusion $\|f\|_{J^{1,p}(X,Y)}\le \|f\|_{W^{1,p}(X\times Y)}$ and show that the norms agree on the algebraic tensor product $W^{1,p}(X)\otimes W^{1,p}(Y)\subset W^{1,p}(X\times Y)$. When $p=2$ and $X$ and $Y$ are infinitesimally quasi-Hilbertian, standard Dirichlet form theory yields the density of $W^{1,2}(X)\otimes W^{1,2}(Y)$ in $J^{1,2}(X,Y)$ thus implying the equality of the spaces. Our approach raises the question of the density of $W^{1,p}(X)\otimes W^{1,p}(Y)$ in $J^{1,p}(X,Y)$ in the general case.