Calibrated Persistent Homology Tests for High-dimensional Collapse Detection

Alexander Kalinowski · 2026

Read Calibrated Persistent Homology Tests for High-dimensional Collapse Detection by Alexander Kalinowski (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2604-26068.txt.

Description

Kalinowski studies how to detect collapse in high-dimensional point clouds, where mass piles up near a lower-dimensional shape instead of spreading out. He builds test statistics from persistent homology, a tool that tracks holes and connected pieces across scales, under two standard filtrations, with cutoffs calibrated against many non-collapsed reference models. Benchmarks across three collapse mechanisms, linear, nonlinear-support, and contamination-driven, yield a map guiding which filtration and statistic to choose.

Catalog record

CREATORAlexander Kalinowski
YEAR2026
COLLECTIONPAPERS
FORMATTXT
FILE SIZE8442
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2604.26068

Browse the full library · About the Tower of Babel