Quantitative stratification and the regularity of mean curvature flow

From MaRDI portal
Publication:2376329



Abstract: Let cM be a Brakke flow of n-dimensional surfaces in RN. The singular set cSsubsetcM has a stratification cS0subsetcS1subset...cS, where XincSj if no tangent flow at X has more than j symmetries. Here, we define quantitative singular strata cSeta,rj satisfying cupeta>0cap0<rcSeta,rj=cSj. Sharpening the known parabolic Hausdorff dimension bound dimcSjleqj, we prove the effective Minkowski estimates that the volume of r-tubular neighborhoods of cSeta,rj satisfies Vol(Tr(cSeta,rj)capB1)leqCrN+2−j−varepsilon. Our primary application of this is to higher regularity of Brakke flows starting at k-convex smooth compact embedded hypersurfaces. To this end, we prove that for the flow of k-convex hypersurfaces, any backwards selfsimilar limit flow with at least k symmetries is in fact a static multiplicity one plane. Then, denoting by cBrsubsetcM the set of points with regularity scale less than r, we prove that Vol(Tr(cBr))leqCrn+4−k−varepsilon. This gives Lp-estimates for the second fundamental form for any p<n+1−k. In fact, the estimates are much stronger and give Lp-estimates for the inverse of the regularity scale. These estimates are sharp. The key technique that we develop and apply is a parabolic version of the quantitative stratification method introduced in Cheeger and Naber (arXiv:1103.1819v3) and Cheeger and Naber (arXiv:1107.3097v1).


This deep technical paper contains several results regarding the regularity of Brakke flows \(\mathcal{M}=\{(M_t, t) \mid t \in I\}\) of \(n\)-dimensional surfaces in \(\mathbb{R}^N\). At every point \(X \in \mathcal{M}\), there exists a tangent flow obtained by blow up at \(X\). While the tangent flow at \(X\) may not be unique, all tangent flows are self-similar. A stratum \(\mathcal{S}^j(\mathcal{M})\) can be defined by the number of symmetries of the tangent flow at each point. Brian White proved some time ago that the Hausdorff dimension of the strata satisfies \(\dim\mathcal{S}^j(\mathcal{M}) \leq j\). The first result of the present paper is a refinement of White's results. The authors introduce a quantitative stratification \(\mathcal{S}_{\eta, r}^j(\mathcal{M})\) which separates the points of \(\mathcal{M}\) based on \textit{almost symmetries} of parabolic balls of \(\mathcal{M}\) of a fixed radius. The standard stratification can be recovered from the new one via \(\bigcup_{\eta >0} \bigcap_{r>0}\mathcal{S}_{\eta, r}^j(\mathcal{M})=\mathcal{S}^j(\mathcal{M}) \). The new result concerns an estimate of the stronger Minkowski dimension of the quantitative strata. The proof builds up on earlier techniques of quantitative stratification developed by Cheeger and Naber adapted here in the parabolic setting. The main applications are two regularity results for the Brakke flow of \(k\)-convex hypersurfaces. Roughly, one application is a new \(\epsilon\)-regularity theorem for the \(k\)-convex mean curvature flows, while the second asserts that away from a set of small volume, the Brakke flow can be written as a smooth single-values graph of definite size with sharp estimates.



Cites work


Cited in
(32)








This page was built for publication: Quantitative stratification and the regularity of mean curvature flow

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2376329)