An isomorphic version of the slicing problem

DOI10.1016/J.JFA.2004.05.003MaRDI QIDQ1763938zbMATH OpenOpenAlexFDO

Authors B. Klartag

Publication date 22 February 2005

Published in Journal of Functional Analysis (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/math/0312475

zbMATH Keywords

Banach--Mazur distance convex body slicing problem

Mathematics Subject Classification ID

Convex sets in (n) dimensions (including convex hypersurfaces) (52A20)

Abstract: Here we show that any n-dimensional centrally symmetric convex body K has an n-dimensional perturbation T which is convex and centrally symmetric, such that the isotropic constant of T is universally bounded. T is close to K in the sense that the Banach-Mazur distance between T and K is O(log n). If K has a non-trivial type then the distance is universally bounded. In addition, if K is quasi-convex then there exists a quasi-convex T with a universally bounded isotropic constant and with a universally bounded distance to K.

Review text

Let \({\mathcal C}({\mathbb R}^n)\) denote the set od all centrally-symmetric convex bodies in~\({\mathbb R}^n\). Given \(K,T\in {\mathcal C}({\mathbb R}^n)\), \(d_{BM}(K,T)\) will denote the Banach-Mazur distance between \(K\) and~\(T\), and \(L_T\) denote the isotropic constant of \(T\). The following relaxation form of the slicing problem is considered: Do there exist \(c_1,c_2>0\) such that for any \(n\in {\mathbb N}\) and for any \(K,T\in {\mathcal C}({\mathbb R}^n)\) there exists \(T\in {\mathcal C}({\mathbb R}^n)\) such that \(d_{BM}(K,T)<c_1\) and \(L_T<c_2\)? In this note the answer is given affirmatively, up to~\(\log n\): For any \(n\in {\mathbb N}\) and any \(K\in {\mathcal C}({\mathbb R}^n)\) there exist \(T\in {\mathcal C}({\mathbb R}^n)\), \(c_1,c_2>0\) such that \(d_{BM}(K,T) <c_1\log n\) and \(L_T<c_2\). A similar answer (uniformly bounded) is obtained for quasi-convex sets.

Recommendations

Cites work

Cited in

(18)

This page was built for publication: An isomorphic version of the slicing problem

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q1763938)