We focus on the problem of manifold estimation: given a set of observations sampled close to some unknown submanifold M, one wants to recover information about the geometry of M. Minimax estimators which have been proposed so far all depend crucially on the a priori knowledge of parameters quantifying the underlying distribution generating the sample (such as bounds on its density), whereas those quantities will be unknown in practice. Our contribution to the matter is twofold. First, we introduce a one-parameter family of manifold estimators based on a localized version of convex hulls, and show that for some choice of t, the corresponding estimator is minimax on the class of models of manifolds introduced in . Second, we propose a completely data-driven selection procedure for the parameter t, leading to a minimax adaptive manifold estimator on this class of models. This selection procedure actually allows us to recover the Hausdorff distance between the set of observations and M, and can therefore be used as a scale parameter in other settings, such as tangent space estimation.
I am grateful to Fréderic Chazal (Inria Saclay) and Pascal Massart (Université Paris-Sud) for thoughtful discussions and valuable comments on both mathematical and computational aspects of this work. I would also like to thank the anonymous reviewers for their helpful suggestions.
"Minimax adaptive estimation in manifold inference." Electron. J. Statist. 15 (2) 5888 - 5932, 2021. https://doi.org/10.1214/21-EJS1934