Banach lattices with upper $p$-estimates: Renorming and factorization

TL;DR

This paper explores Banach lattices with upper p-estimates, extending classical $p$-convexity results, developing new tools, and revealing fundamental differences in their structure and renorming properties.

Contribution

It provides a comprehensive analysis of Banach lattices with upper p-estimates, including representation, factorization, and structural theorems, using novel methods avoiding convexification.

Findings

Many classical $p$-convexity theorems have analogues for upper $p$-estimates.

All such spaces can be represented inside infinity sums of model spaces.

New tools are developed to analyze these spaces without convexification.

Abstract

The notions of $p$-convexity and concavity are fundamental tools for studying Banach lattices, as they partition the class of Banach lattices into a scale of spaces with $L_p$-like properties. Upper and lower $p$-estimates provide a refinement of this scale, modeled by the Lorentz spaces $L_{p,\infty}$ and $L_{p,1}$, respectively. In this article, we provide a comprehensive treatment of Banach lattices with upper $p$-estimates. In particular, we show that many well-known theorems about $p$-convex Banach lattices have analogues in the upper $p$-estimate setting, including the ability to represent all such spaces inside of infinity sums of model spaces, to canonically factor the convex operators and identify their associated operator ideals, as well as to give a precise description of the free objects and push-outs. Proving these results is far from straightforward and will require the development of a variety of new tools that avoid convexification and concavification procedures. In fact, we will identify many fundamental differences between the theories of $p$-convexity and upper $p$-estimates, particularly with regards to isometric problems and renormings.