English

The John inclusion for log-concave functions

Metric Geometry 2026-01-16 v2

Abstract

John's inclusion states that a convex body in Rd\mathbb{R}^d can be covered by the dd-dilation of its maximal volume ellipsoid. We obtain a certain John-type inclusion for log-concave functions. As a byproduct of our approach, we establish the following asymptotically tight inequality: \\ \noindent For any log-concave function ff with finite, positive integral, there exist a positive definite matrix AA, a point aRda \in \mathbb{R}^d, and a positive constant α\alpha such that χBd(x)αf ⁣ ⁣(A(xa))d+1exd+2+(d+1), \chi_{\mathbf{B}^{d}}(x) \leq \alpha f\!\!\left(A(x-a)\right) \leq \sqrt{d+1} \cdot e^{-\frac{\left|x\right|}{d+2} + (d+1)}, where χBd\chi_{\mathbf{B}^{d}} is the indicator function of the unit ball Bd\mathbf{B}^{d}.

Keywords

Cite

@article{arxiv.2412.18444,
  title  = {The John inclusion for log-concave functions},
  author = {G. Ivanov},
  journal= {arXiv preprint arXiv:2412.18444},
  year   = {2026}
}
R2 v1 2026-06-28T20:48:06.225Z