Saved in:
Bibliographic Details
Main Authors: Green, Dylan, Bailey, Stephen
Format: Preprint
Published: 2023
Subjects:
Online Access:https://arxiv.org/abs/2311.04855
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866929525091205120
author Green, Dylan
Bailey, Stephen
author_facet Green, Dylan
Bailey, Stephen
contents Non-negative matrix factorization (NMF) is a dimensionality reduction technique that has shown promise for analyzing noisy data, especially astronomical data. For these datasets, the observed data may contain negative values due to noise even when the true underlying physical signal is strictly positive. Prior NMF work has not treated negative data in a statistically consistent manner, which becomes problematic for low signal-to-noise data with many negative values. In this paper we present two algorithms, Shift-NMF and Nearly-NMF, that can handle both the noisiness of the input data and also any introduced negativity. Both of these algorithms use the negative data space without clipping, and correctly recover non-negative signals without any introduced positive offset that occurs when clipping negative data. We demonstrate this numerically on both simple and more realistic examples, and prove that both algorithms have monotonically decreasing update rules.
format Preprint
id arxiv_https___arxiv_org_abs_2311_04855
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle Algorithms for Non-Negative Matrix Factorization on Noisy Data With Negative Values
Green, Dylan
Bailey, Stephen
Instrumentation and Methods for Astrophysics
Machine Learning
Signal Processing
Methodology
Non-negative matrix factorization (NMF) is a dimensionality reduction technique that has shown promise for analyzing noisy data, especially astronomical data. For these datasets, the observed data may contain negative values due to noise even when the true underlying physical signal is strictly positive. Prior NMF work has not treated negative data in a statistically consistent manner, which becomes problematic for low signal-to-noise data with many negative values. In this paper we present two algorithms, Shift-NMF and Nearly-NMF, that can handle both the noisiness of the input data and also any introduced negativity. Both of these algorithms use the negative data space without clipping, and correctly recover non-negative signals without any introduced positive offset that occurs when clipping negative data. We demonstrate this numerically on both simple and more realistic examples, and prove that both algorithms have monotonically decreasing update rules.
title Algorithms for Non-Negative Matrix Factorization on Noisy Data With Negative Values
topic Instrumentation and Methods for Astrophysics
Machine Learning
Signal Processing
Methodology
url https://arxiv.org/abs/2311.04855