Proof of the log-convex density conjecture (Q2319578)
From MaRDI portal
scientific article
Language | Label | Description | Also known as |
---|---|---|---|
English | Proof of the log-convex density conjecture |
scientific article |
Statements
Proof of the log-convex density conjecture (English)
0 references
20 August 2019
0 references
For a set \(A\subset\mathbb{R}^n\) of locally finite perimeter and a positive function (density) \(f\) on \(\mathbb{R}^n\), the weighted perimeter and weighted volume of \(A\) are defined respectively as \[ \operatorname{Per}(A)=\int_{\partial A}f\, d\mathcal{H}^{n-1} \text{ and } \mathrm{Vol}(A)=\int_{A}f\, d\mathcal{H}^{n}, \] where \(\mathcal{H}^{m}\) is the \(m\)-dimensional Hausdorff measure, and \(\partial A\) is the essential boundary of \(A\). If \(M>0\) is a fixed volume, then the question arises whether there exists an isoperimetric set \(A\subset\mathbb{R}^n\) such that \(\operatorname{Vol}(A)=M\) and \(\operatorname{Per}(A)=\inf\limits_{\substack {Q\subset\mathbb{R}^n\\ \operatorname{Vol}(Q)=M }}\operatorname{Per}(Q)\). In [Calc. Var. Partial Differ. Equ. 31, No. 1, 27--46 (2008; Zbl 1126.49038)], \textit{C. Rosales} et al. studied this problem for radial log-convex densities of the form \(f(x)=e^{g(|x|)}\), where \(g\) is a smooth, convex, and even function on \(\mathbb{R}\), and they posed the conjecture stating that in \(\mathbb{R}^n\) with a smooth, radial, log-convex density, balls around the origin provide isoperimetric regions of any given volume. In this paper, the author completely solves this conjecture by showing that if \(f(x)=e^{g(|x|)}\) is a density on \(\mathbb{R}^n\) with a smooth, convex, and even function \(g\), then balls around the origin are isoperimetric regions with respect to weighted perimeter and volume. Moreover, the author proves the uniqueness of isoperimetric regions by showing that up to sets of measure \(0\), the only isoperimetric regions are balls centered at the origin, and balls that lie entirely in \(B_{\mathcal{R}(f)}=\{x;\ |x|\le\mathcal{R}(f)\}\), where \(\mathcal{R}(f)=\sup\{|x|;\ f(x)=f(0)\}\).
0 references
isoperimetric regions
0 references
log-convex density
0 references