ResearchTrend.AI
  • Papers
  • Communities
  • Events
  • Blog
  • Pricing
Papers
Communities
Social Events
Terms and Conditions
Pricing
Parameter LabParameter LabTwitterGitHubLinkedInBlueskyYoutube

© 2025 ResearchTrend.AI, All rights reserved.

  1. Home
  2. Papers
  3. 2108.03135
20
9

Minimax Boundary Estimation and Estimation with Boundary

6 August 2021
Eddie Aamari
C. Aaron
Clément Levrard
ArXivPDFHTML
Abstract

We derive non-asymptotic minimax bounds for the Hausdorff estimation of ddd-dimensional submanifolds M⊂RDM \subset \mathbb{R}^DM⊂RD with (possibly) non-empty boundary ∂M\partial M∂M. The model reunites and extends the most prevalent C2\mathcal{C}^2C2-type set estimation models: manifolds without boundary, and full-dimensional domains. We consider both the estimation of the manifold MMM itself and that of its boundary ∂M\partial M∂M if non-empty. Given nnn samples, the minimax rates are of order O((log⁡n/n)2/d)O\bigl((\log n/n)^{2/d}\bigr)O((logn/n)2/d) if ∂M=∅\partial M = \emptyset∂M=∅ and O((log⁡n/n)2/(d+1))O\bigl((\log n/n)^{2/(d+1)}\bigr)O((logn/n)2/(d+1)) if ∂M≠∅\partial M \neq \emptyset∂M=∅, up to logarithmic factors. In the process, we develop a Voronoi-based procedure that allows to identify enough points O((log⁡n/n)2/(d+1))O\bigl((\log n/n)^{2/(d+1)}\bigr)O((logn/n)2/(d+1))-close to ∂M\partial M∂M for reconstructing it.

View on arXiv
Comments on this paper