Personalized Tucker Decomposition: Modeling Commonality and Peculiarity on Tensor Data

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Hu, Jiuyun, Shi, Naichen, Kontar, Raed Al, Yan, Hao
Natura: Preprint
Pubblicazione: 2023
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866916911503114240
author Hu, Jiuyun
Shi, Naichen
Kontar, Raed Al
Yan, Hao
author_facet Hu, Jiuyun
Shi, Naichen
Kontar, Raed Al
Yan, Hao
contents We propose personalized Tucker decomposition (perTucker) to address the limitations of traditional tensor decomposition methods in capturing heterogeneity across different datasets. perTucker decomposes tensor data into shared global components and personalized local components. We introduce a mode orthogonality assumption and develop a proximal gradient regularized block coordinate descent algorithm that is guaranteed to converge to a stationary point. By learning unique and common representations across datasets, we demonstrate perTucker's effectiveness in anomaly detection, client classification, and clustering through a simulation study and two case studies on solar flare detection and tonnage signal classification.
format Preprint
id arxiv_https___arxiv_org_abs_2309_03439
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle Personalized Tucker Decomposition: Modeling Commonality and Peculiarity on Tensor Data
Hu, Jiuyun
Shi, Naichen
Kontar, Raed Al
Yan, Hao
Machine Learning
Methodology
We propose personalized Tucker decomposition (perTucker) to address the limitations of traditional tensor decomposition methods in capturing heterogeneity across different datasets. perTucker decomposes tensor data into shared global components and personalized local components. We introduce a mode orthogonality assumption and develop a proximal gradient regularized block coordinate descent algorithm that is guaranteed to converge to a stationary point. By learning unique and common representations across datasets, we demonstrate perTucker's effectiveness in anomaly detection, client classification, and clustering through a simulation study and two case studies on solar flare detection and tonnage signal classification.
title Personalized Tucker Decomposition: Modeling Commonality and Peculiarity on Tensor Data
topic Machine Learning
Methodology
url https://arxiv.org/abs/2309.03439