A parametrized compactness theorem under bounded Ricci curvature (Q1705058): Difference between revisions

The authors study the parametrized compactness of non-collapsed manifolds of bounded Ricci curvature with diameter bounded from above and injectivity radius bounded from below in the following form (see Theorem A): Given \(m,\lambda,\mu,\rho >0\), there are constants \(\delta_0 (m, \lambda, \mu, \rho) > 0\) and \(\varepsilon_0 (m, \lambda, \mu, \rho) > 0\), such that for any \(\delta \in [0, \delta_0 )\) and \(\varepsilon \in (0, \varepsilon_0 )\), if two elements \(f_i : (M_i, g_i)\to (N_i , h_i)\in{\mathcal F}_{\mathrm{Ric}}^{m,n}(\delta,\lambda,\mu,\rho,D)\), \(i=1,2\), are \(\varepsilon\)-close in the Gromov-Hausdorff topology, then there exists a pair of diffeomorphisms \((\Phi,\Psi)\) such that \(\Phi : (M_1 , g_1 ) \to (M_2, g_2)\) is \(e^{\varkappa(\delta,\varepsilon)}\)-bi-Lipschitz, \(\Psi : (N_1 , h_1) \to (N_2 , h_2 )\) is \(e^{\varkappa(\delta,\varepsilon)}\)-bi-Lipschitz, and \(\Psi \circ f_1 = f_2 \circ \Phi\), where \(\varkappa(\delta,\varepsilon)=\varkappa(\delta,\varepsilon\mid m,\lambda,\mu,\rho)\) and \(\varkappa(a, b | c, \dots )\) denotes a positive function in \(a, b, c,\dots\) such that after fixing the other parameters, \(\varkappa(a, b) = \varkappa(a, b\mid c, \dots) \to 0\) as \(a, b \to 0\). This theorem generalizes, among others, the main result in [\textit{M. T. Anderson}, Invent. Math. 102, No. 2, 429--445 (1990; Zbl 0711.53038)].