Skip to content

Latest commit

 

History

History
59 lines (59 loc) · 2.29 KB

File metadata and controls

59 lines (59 loc) · 2.29 KB
title Non-Oblivious Performance of Random Projections
booktitle Proceedings of the 16th Asian Conference on Machine Learning
year 2025
volume 260
series Proceedings of Machine Learning Research
month 0
publisher PMLR
pdf https://raw.githubusercontent.com/mlresearch/v260/main/assets/skorski25a/skorski25a.pdf
url https://proceedings.mlr.press/v260/skorski25a.html
software https://github.com/maciejskorski/NonobliviousRademacherProjections/
openreview 9XBg73M0BE
abstract Random projections are a cornerstone of high-dimensional computations. However, their analysis has proven both difficult and inadequate in capturing the empirically observed accuracy. To bridge this gap, this paper studies random projections from a novel perspective, focusing on data-dependent, that is, \emph{non-oblivious}, performance. The key contribution is the precise and data-dependent accuracy analysis of Rademacher random projections, achieved through elegant geometric methods of independent interest, namely, \emph{Schur-concavity}. The result formally states the following property: the less spread-out the data is, the better the accuracy. This leads to notable improvements in accuracy guarantees for data characterized by sparsity or distributed with a small spread. The key tool is a novel algebraic framework for proving Schur-concavity properties, which offers an alternative to derivative-based criteria commonly used in related studies. We demonstrate its value by providing an alternative proof for the extension of the celebrated Khintchine inequality.
layout inproceedings
issn 2640-3498
id skorski25a
tex_title Non-Oblivious Performance of Random Projections
firstpage 1128
lastpage 1143
page 1128-1143
order 1128
cycles false
bibtex_editor Nguyen, Vu and Lin, Hsuan-Tien
editor
given family
Vu
Nguyen
given family
Hsuan-Tien
Lin
bibtex_author Skorski, Maciej and Temperoni, Alessandro
author
given family
Maciej
Skorski
given family
Alessandro
Temperoni
date 2025-01-14
address
container-title Proceedings of the 16th Asian Conference on Machine Learning
genre inproceedings
issued
date-parts
2025
1
14
extras