A generalizable framework for low-rank tensor completion with numerical priors

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Yuan, Shiran, Huang, Kaizhu
Format: Preprint
Veröffentlicht: 2023
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866929390475018240
author Yuan, Shiran
Huang, Kaizhu
author_facet Yuan, Shiran
Huang, Kaizhu
contents Low-Rank Tensor Completion, a method which exploits the inherent structure of tensors, has been studied extensively as an effective approach to tensor completion. Whilst such methods attained great success, none have systematically considered exploiting the numerical priors of tensor elements. Ignoring numerical priors causes loss of important information regarding the data, and therefore prevents the algorithms from reaching optimal accuracy. Despite the existence of some individual works which consider ad hoc numerical priors for specific tasks, no generalizable frameworks for incorporating numerical priors have appeared. We present the Generalized CP Decomposition Tensor Completion (GCDTC) framework, the first generalizable framework for low-rank tensor completion that takes numerical priors of the data into account. We test GCDTC by further proposing the Smooth Poisson Tensor Completion (SPTC) algorithm, an instantiation of the GCDTC framework, whose performance exceeds current state-of-the-arts by considerable margins in the task of non-negative tensor completion, exemplifying GCDTC's effectiveness. Our code is open-source.
format Preprint
id arxiv_https___arxiv_org_abs_2302_05881
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle A generalizable framework for low-rank tensor completion with numerical priors
Yuan, Shiran
Huang, Kaizhu
Computer Vision and Pattern Recognition
Artificial Intelligence
Machine Learning
Numerical Analysis
Low-Rank Tensor Completion, a method which exploits the inherent structure of tensors, has been studied extensively as an effective approach to tensor completion. Whilst such methods attained great success, none have systematically considered exploiting the numerical priors of tensor elements. Ignoring numerical priors causes loss of important information regarding the data, and therefore prevents the algorithms from reaching optimal accuracy. Despite the existence of some individual works which consider ad hoc numerical priors for specific tasks, no generalizable frameworks for incorporating numerical priors have appeared. We present the Generalized CP Decomposition Tensor Completion (GCDTC) framework, the first generalizable framework for low-rank tensor completion that takes numerical priors of the data into account. We test GCDTC by further proposing the Smooth Poisson Tensor Completion (SPTC) algorithm, an instantiation of the GCDTC framework, whose performance exceeds current state-of-the-arts by considerable margins in the task of non-negative tensor completion, exemplifying GCDTC's effectiveness. Our code is open-source.
title A generalizable framework for low-rank tensor completion with numerical priors
topic Computer Vision and Pattern Recognition
Artificial Intelligence
Machine Learning
Numerical Analysis
url https://arxiv.org/abs/2302.05881