arrow
返回

Scalable tensor factorizations for incomplete data

delete2011-03-01
delete503
delete
OA
AI
E
Evrim Acar
D
Daniel Dunlavy
T
Tamara G. Kolda *
M
Morten Mørup
DOI:10.1016/j.chemolab.2010.08.004delete
delete原文链接
delete分享
delete收藏
查看原文
摘要

摘要

En 中文
The problem of incomplete data - i.e., data with missing or unknown values - in multi-way arrays is ubiquitous in biomedical signal processing, network traffic analysis, bibliometrics, social network analysis, chemometrics, computer vision, communication networks, etc. We consider the problem of how to factorize data sets with missing values with the goal of capturing the underlying latent structure of the data and possibly reconstructing missing values (i.e., tensor completion). We focus on one of the most well-known tensor factorizations that captures multi-linear structure, CANDECOMP/PARAFAC (CP). In the presence of missing data, CP can be formulated as a weighted least squares problem that models only the known entries. We develop an algorithm called CP-WOPT (CP Weighted OPTimization) that uses a first-order optimization approach to solve the weighted least squares problem. Based on extensive numerical experiments, our algorithm is shown to successfully factorize tensors with noise and up to 99% missing data. A unique aspect of our approach is that it scales to sparse large-scale data, e.g., 1000 x 1000 x 1000 with five million known entries (0.5% dense). We further demonstrate the usefulness of CP-WOPT on two real-world applications: a novel EEG (electroencephalogram) application where missing data is frequently encountered due to disconnections of electrodes and the problem of modeling computer network traffic where data may be absent due to the expense of the data collection process. (C) 2010 Elsevier B.V. All rights reserved.
Keyword:
Missing data
Incomplete data
Tensor factorization
CANDECOMP
PARAFAC
Optimization
AI总结

AI总结

对已上传原文的论文进行重点信息的提取,主要内容包括:简要概述、研究摘要、背景介绍、关键亮点、图文解析、展望与总结。

期刊

Chemometrics and Intelligent Laboratory Systems 封面图
Chemometrics and Intelligent Laboratory Systems
IF:
3.8
论文数:
4.6K
被引数:
1.2W

机构

T
turkiye bilimsel ve teknolojik arastirma kurumu (tubitak)
学者数:
1.4K
论文数: 1.3K
被引数: 3
U
united states department of energy (doe)
学者数:
11.3W
论文数: 9.6W
被引数: 246
S
Sandia National Laboratories
学者数:
5.5K
论文数: 3.8K
被引数: 6.4K
学者 查看更多机构
引用论文

引用论文

Prevalent and persistent viral infection in cultures of the coral algal endosymbiont Symbiodinium
err2017-03-17
err0
PREAI
errKaren D. Weynberg; Matthew Neave; Peta L. Clode; Christian R. Voolstra; Christopher Brownlee; Patrick Laffy; Nicole S. Webster; Rachel A. Levin; Elisha M. Wood-Charlson; Madeleine J. H. van Oppen
err分享
err收藏
err分享
err收藏
Structural, optical and magnetic properties of Co-doped ZnO nanorods with hidden secondary phases
err2008-10-09
err0
PREAI
errXuefeng Wang; Rongkun Zheng; Zongwen Liu; Ho-pui Ho; Jianbin Xu; Simon P Ringer
err分享
err收藏
Calcium-Binding Capacity of Centrin2 Is Required for Linear POC5 Assembly but Not for Nucleotide Excision Repair
err2013-07-02
err0
errOAAI
errTiago J. Dantas; Owen M. Daly; Pauline C. Conroy; Martin Tomas; Yifan Wang; Pierce Lalor; Peter Dockery; Elisa Ferrando-May; Ciaran G. Morrison
err分享
err收藏
Tensor Decompositions and Applications张量分解及其应用
err2009-08-05
err7.5K
PREAI
errKolda, Tamara G.; Bader, Brett W.
err分享
err收藏
err分享
err收藏
err分享
err收藏
学者 查看更多内容