Stratified Non-Negative Tensor Factorization

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Sietsema, Alexander, Vural, Zerrin, Chapman, James, Yaniv, Yotam, Needell, Deanna
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866912136259698688
author Sietsema, Alexander
Vural, Zerrin
Chapman, James
Yaniv, Yotam
Needell, Deanna
author_facet Sietsema, Alexander
Vural, Zerrin
Chapman, James
Yaniv, Yotam
Needell, Deanna
contents Non-negative matrix factorization (NMF) and non-negative tensor factorization (NTF) decompose non-negative high-dimensional data into non-negative low-rank components. NMF and NTF methods are popular for their intrinsic interpretability and effectiveness on large-scale data. Recent work developed Stratified-NMF, which applies NMF to regimes where data may come from different sources (strata) with different underlying distributions, and seeks to recover both strata-dependent information and global topics shared across strata. Applying Stratified-NMF to multi-modal data requires flattening across modes, and therefore loses geometric structure contained implicitly within the tensor. To address this problem, we extend Stratified-NMF to the tensor setting by developing a multiplicative update rule and demonstrating the method on text and image data. We find that Stratified-NTF can identify interpretable topics with lower memory requirements than Stratified-NMF. We also introduce a regularized version of the method and demonstrate its effects on image data.
format Preprint
id arxiv_https___arxiv_org_abs_2411_18805
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Stratified Non-Negative Tensor Factorization
Sietsema, Alexander
Vural, Zerrin
Chapman, James
Yaniv, Yotam
Needell, Deanna
Machine Learning
Numerical Analysis
G.1.6; I.5.3; I.5.4
Non-negative matrix factorization (NMF) and non-negative tensor factorization (NTF) decompose non-negative high-dimensional data into non-negative low-rank components. NMF and NTF methods are popular for their intrinsic interpretability and effectiveness on large-scale data. Recent work developed Stratified-NMF, which applies NMF to regimes where data may come from different sources (strata) with different underlying distributions, and seeks to recover both strata-dependent information and global topics shared across strata. Applying Stratified-NMF to multi-modal data requires flattening across modes, and therefore loses geometric structure contained implicitly within the tensor. To address this problem, we extend Stratified-NMF to the tensor setting by developing a multiplicative update rule and demonstrating the method on text and image data. We find that Stratified-NTF can identify interpretable topics with lower memory requirements than Stratified-NMF. We also introduce a regularized version of the method and demonstrate its effects on image data.
title Stratified Non-Negative Tensor Factorization
topic Machine Learning
Numerical Analysis
G.1.6; I.5.3; I.5.4
url https://arxiv.org/abs/2411.18805