A theory of stratification learning

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Aamari, Eddie, Berenfeld, Clément
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866929366304292864
author Aamari, Eddie
Berenfeld, Clément
author_facet Aamari, Eddie
Berenfeld, Clément
contents Given i.i.d. sample from a stratified mixture of immersed manifolds of different dimensions, we study the minimax estimation of the underlying stratified structure. We provide a constructive algorithm allowing to estimate each mixture component at its optimal dimension-specific rate adaptively. The method is based on an ascending hierarchical co-detection of points belonging to different layers, which also identifies the number of layers and their dimensions, assigns each data point to a layer accurately, and estimates tangent spaces optimally. These results hold regardless of any ambient assumption on the manifolds or on their intersection configurations. They open the way to a broad clustering framework, where each mixture component models a cluster emanating from a specific nonlinear correlation phenomenon.
format Preprint
id arxiv_https___arxiv_org_abs_2405_20066
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle A theory of stratification learning
Aamari, Eddie
Berenfeld, Clément
Statistics Theory
Differential Geometry
62G05, 57R42, 54E20
Given i.i.d. sample from a stratified mixture of immersed manifolds of different dimensions, we study the minimax estimation of the underlying stratified structure. We provide a constructive algorithm allowing to estimate each mixture component at its optimal dimension-specific rate adaptively. The method is based on an ascending hierarchical co-detection of points belonging to different layers, which also identifies the number of layers and their dimensions, assigns each data point to a layer accurately, and estimates tangent spaces optimally. These results hold regardless of any ambient assumption on the manifolds or on their intersection configurations. They open the way to a broad clustering framework, where each mixture component models a cluster emanating from a specific nonlinear correlation phenomenon.
title A theory of stratification learning
topic Statistics Theory
Differential Geometry
62G05, 57R42, 54E20
url https://arxiv.org/abs/2405.20066