{"id":1387,"date":"2013-04-24T05:57:09","date_gmt":"2013-04-24T10:57:09","guid":{"rendered":"http:\/\/sites.williams.edu\/Morgan\/?p=1387"},"modified":"2013-05-04T06:05:51","modified_gmt":"2013-05-04T11:05:51","slug":"isoperimetric-regions-with-density","status":"publish","type":"post","link":"https:\/\/sites.williams.edu\/Morgan\/2013\/04\/24\/isoperimetric-regions-with-density\/","title":{"rendered":"Isoperimetric Regions with Density"},"content":{"rendered":"<p>In\u00a0<strong>R<\/strong><sup><em>n <\/em><\/sup>or in a Riemannian manifold, one may consider regions <em>R<\/em> with density given by an integrable nonnegative function <em>g<\/em>, with volume \u222b<em><sub>R\u00a0<\/sub>g<\/em>. If everything is smooth, the perimeter is given by \u222b<em><sub>\u2202R\u00a0<\/sub>g<\/em>, or more generally by Stokes&#8217; Theorem. For finite perimeter, these are the so-called normal currents of geometric measure theory. All of this can be done in a <a href=\"http:\/\/www.ams.org\/notices\/200508\/fea-morgan.pdf\">manifold with density <em>f<\/em><\/a>\u00a0(unrelated to <em>g<\/em>).<\/p>\n<p>In\u00a0<strong>R<\/strong><sup><em>n<\/em><\/sup>, if you allow regions with density, there is no isoperimetric optimum for given volume, because large balls with low (constant) densities approach perimeter 0; similarly in any space for which <em>P<\/em>\/<em>V<\/em> has no minimum. In a space of finite volume, such as Gauss space G<sup><em>n<\/em><\/sup>\u00a0(<strong>R<\/strong><sup><em>n<\/em><\/sup> with Gaussian density) there is an optimum: the whole space with appropriate constant density has perimeter 0. At the other extreme, in <strong>R<\/strong> with density exp(<em>x<\/em><sup>3<\/sup>), a left halfline with high density approaches perimeter 0.<\/p>\n<p>Note that if a region with density is isoperimetric, it remains so for all multiples of that density.<\/p>\n<p>In principle one may allow regions with variable density <em>f<\/em>, the variability contributing to the boundary, but since any such region is an integral of regions <em>R<sub>d<\/sub><\/em> = {<em>f<\/em> \u2265 <em>d<\/em>} with constant density <em>d<\/em>, one need consider only regions with constant density.<\/p>\n<p>For a manifold with density with isoperimetric profile <em>P<\/em> = <em>I<\/em>(<em>V<\/em>), the least-perimeter region with density with unit volume just minimizes <em>cI<\/em>(1\/<em>c<\/em>); in a space of infinite volume its perimeter to volume ratio is the <a href=\"http:\/\/en.wikipedia.org\/wiki\/Cheeger_constant\">Cheeger constant<\/a> inf <em>P<\/em>\/<em>V<\/em>.<\/p>\n<p>In <strong>R<\/strong><sup>2<\/sup>\u00a0with density exp(<em>r<\/em><sup>2<\/sup>), where <a href=\"http:\/\/sites.williams.edu\/Morgan\/2010\/04\/03\/the-log-convex-density-conjecture\/\">classical isoperimetric regions are balls about the origin<\/a>, among regions with density the isoperimetric optimum for every area is the same disc of radius r ~ 1.1 (solution to exp(<em>r<\/em><sup>2<\/sup>)\u00a0= 2<em>r<\/em><sup>2<\/sup> + 1) with appropriate constant density.<\/p>\n<p><strong>Proposition.<\/strong> <em>In<\/em> <strong>R<\/strong><sup><em>n<\/em><\/sup>\u00a0<em>with\u00a0smooth,\u00a0uniformly log-convex radial density f(r),<\/em> <em>an isoperimetric region with density exists.<\/em><\/p>\n<p><em><!--more-->Proof<\/em>. By <a href=\"http:\/\/arxiv.org\/abs\/1111.5160\">Morgan-Pratelli<\/a> (Thms. 3.3, 5.8), a classical, compact \u00a0region\u00a0<em>R<\/em>\u00a0of least perimeter <em>P<\/em> exists for every given volume <em>V<\/em>. By Kolesnikov-Zhdanov [ZD] (see also\u00a0<a href=\"http:\/\/arxiv.org\/abs\/1107.4402\">Howe<\/a>\u00a0Cor. 3.7) and Figalli-Maggi [FM], for large and small volumes\u00a0<em>R<\/em> is a ball about the origin. Consequently, as <em>V<\/em> approaches 0 or infinity, <em>P<\/em>\/<em>V\u00a0<\/em>goes to infinity. (For small balls this is trivial. For large balls it follows from the fact that d<em>P<\/em>\/d<em>V<\/em> = (1\/<em>P<\/em>)d<em>P<\/em>\/d<em>r<\/em> \u2265 (1\/<em>f<\/em>)d<em>f<\/em>\/d<em>r<\/em> = d(log <em>f<\/em>)\/d<em>r<\/em> goes to infinity.)\u00a0Since <em>P<\/em>\u00a0and hence <em>P\/V<\/em> are continuous functions of <em>V<\/em>, <em>P<\/em>\/<em>V<\/em> attains a minimum, providing an isoperimetric region with density.<\/p>\n<p>[FM] \u00a0A. Figalli and F. Maggi, On the isoperimetric problem for radial log-convex densities,\u00a0Calc. Var. Partial Differential Equations, to appear.<\/p>\n<p>[KD] Alexander V. Kolesnikov and Roman I. Zhdanov, <a href=\"http:\/\/ams.rice.edu\/mathscinet\/search\/publdoc.html?arg3=&amp;co4=AND&amp;co5=AND&amp;co6=AND&amp;co7=AND&amp;dr=all&amp;pg4=AUCN&amp;pg5=AUCN&amp;pg6=PC&amp;pg7=ALLF&amp;pg8=ET&amp;r=1&amp;review_format=html&amp;s4=Kolesnikov&amp;s5=Zhdanov&amp;s6=&amp;s7=&amp;s8=All&amp;vfpref=html&amp;yearRangeFirst=&amp;yearRangeSecond=&amp;yrop=eq\">On isoperimetric sets of radially symmetric measures<\/a>,\u00a0Concentration, functional inequalities and isoperimetry, 123\u2013154, Contemp. Math. 545, Amer. Math. Soc., Providence, RI, 2011; <a href=\"http:\/\/arxiv.org\/abs\/1002.1829v4\">arXiv <\/a>(2010), Cor. 6.8.<\/p>\n<p><em>Remark.<\/em> In\u00a0<strong>R<\/strong>\u00a0with density exp(<em>x<\/em>), for given volume, left halflines are classically isoperimetric and halflines with constant density all have the same perimeter, as do their integral averages, yielding lots of isoperimetric optima.<\/p>\n","protected":false},"excerpt":{"rendered":"<p>In\u00a0Rn or in a Riemannian manifold, one may consider regions R with density given by an integrable nonnegative function g, with volume \u222bR\u00a0g. If everything is smooth, the perimeter is given by \u222b\u2202R\u00a0g, or more generally by Stokes&#8217; Theorem. For finite perimeter, these are the so-called normal currents of geometric measure theory. All of this [&hellip;]<\/p>\n","protected":false},"author":269,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"_acf_changed":false,"footnotes":""},"categories":[14043],"tags":[],"class_list":["post-1387","post","type-post","status-publish","format-standard","hentry","category-math"],"acf":[],"_links":{"self":[{"href":"https:\/\/sites.williams.edu\/Morgan\/wp-json\/wp\/v2\/posts\/1387","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/sites.williams.edu\/Morgan\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/sites.williams.edu\/Morgan\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/sites.williams.edu\/Morgan\/wp-json\/wp\/v2\/users\/269"}],"replies":[{"embeddable":true,"href":"https:\/\/sites.williams.edu\/Morgan\/wp-json\/wp\/v2\/comments?post=1387"}],"version-history":[{"count":21,"href":"https:\/\/sites.williams.edu\/Morgan\/wp-json\/wp\/v2\/posts\/1387\/revisions"}],"predecessor-version":[{"id":1390,"href":"https:\/\/sites.williams.edu\/Morgan\/wp-json\/wp\/v2\/posts\/1387\/revisions\/1390"}],"wp:attachment":[{"href":"https:\/\/sites.williams.edu\/Morgan\/wp-json\/wp\/v2\/media?parent=1387"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/sites.williams.edu\/Morgan\/wp-json\/wp\/v2\/categories?post=1387"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/sites.williams.edu\/Morgan\/wp-json\/wp\/v2\/tags?post=1387"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}