High-dimensional Clustering and Signal Recovery under Block Signals

Wu Su · 2025

Read High-dimensional Clustering and Signal Recovery under Block Signals by Wu Su (2025) 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-2504-08332.txt.

Description

Wu Su and Yumou Qiu study clustering and signal recovery when data have very many features, in this 2025 paper. Their setting assumes the signal comes in blocks, runs of related features that light up together. Standard methods that ignore the block shape need stronger signals to succeed.

The authors propose two methods: one that gathers features block by block for sparse signals, and one based on moving averages for dense signals. Both adapt to uneven noise and tangled correlations in the data. They also prove matching lower bounds that mark where no method, or no fast method, can succeed at all. Their methods reach those fast-method limits in both regimes. Tests on made-up data and on global temperature records back the theory.

Catalog record

CREATORWu Su
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE69 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2504-08332

Browse the full library · About the Tower of Babel