Angular equivalence of normed spaces

Abstract: Angular equivalence is introduced and shown to be an equivalence relation among the norms on a fixed real vector space. It is a finer notion than the usual (topological) notion of norm equivalence. Angularly equivalent norms share certain geometric properties: A norm that is angularly equivalent to a uniformly convex norm is itself uniformly convex. The same is true for strict convexity. Extreme points of the unit balls of angularly equivalent norms occur on the same rays, and if one unit ball is a polyhedron so is the other. Among norms arising from inner products, two norms are angularly equivalent if and only if they are topological equivalent. But, unlike topological equivalence, angular equivalence is able to distinguish between different norms on a finite-dimensional space. In particular, no two

e l l^{p}

norms on

m a t h b b R^{n}

are angularly equivalent.

Review text

The authors introduce a different notion of equivalence of norms, preserving certain properties of the norm that simple topological equivalence does not. For this task, they say that two norms are angularly equivalent if, over all pairs of non-zero vectors, the angle between the pair, determined by one norm, is comparable to the angle between the same pair, determined by the other norm. Here, the abstract notion of angle between vectors \(x,y\) in a normed space \(X\) is given by \(\theta (x,y)=\frac{g(x,y) }{\left\| x\right\| \left\| y\right\| }\) for a certain special function \(g:X\times X\rightarrow [ 0,\infty )\). One of the main results of the paper under review asserts that for real vector spaces, angularly equivalent norms are also topologically equivalent. The proof is quite technical and, as the authors remark, it is possible that a simpler proof will arise in the future. It is interesting to note that, unlike topological equivalence, norms on a finite-dimensional space are not all angularly equivalent. Five interesting open problems are presented at the end of the paper. For instance, do angularly equivalent norms induce angularly equivalent norms on quotient spaces? Reviewer's comments: The proof of Theorem 2.1 contains an unsupported statement. A corrected proof is presented in a corrigendum published by the authors in [J. Math. Anal. Appl. 458, No. 2, 1361--1362 (2018; Zbl 1472.46013)].

Recommendations

Cites work

Cited in

(5)

This page was built for publication: Angular equivalence of normed spaces

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2360034)