Imputation of streaming low-rank tensor data

Morteza Mardani, Gonzalo Mateos, Georgios B Giannakis

Research output: Chapter in Book/Report/Conference proceedingConference contribution

2 Scopus citations

Abstract

Unraveling latent structure by means of multilinear models of tensor data is of paramount importance in timely inference tasks encountered with 'Big Data' analytics. However, increasingly noisy, heterogeneous, and incomplete datasets as well as the need for real-time processing of streaming data pose major challenges to this end. The present paper introduces a novel online (adaptive) algorithm to decompose low-rank tensors with missing entries, and perform imputation as a byproduct. The novel estimator minimizes an exponentially- weighted least-squares fitting error along with a separable regularizer of the PARAFAC decomposition factors, to trade-off fidelity for complexity of the approximation captured by the decomposition's rank. Leveraging stochastic gradient descent iterations, a scalable, real-time algorithm is developed and its convergence is established under simplifying technical assumptions. Simulated tests with cardiac magnetic resonance imagery (MRI) data confirm the efficacy of the proposed algorithm in imputing up to 75% missing entries.

Original languageEnglish (US)
Title of host publication2014 IEEE 8th Sensor Array and Multichannel Signal Processing Workshop, SAM 2014
PublisherIEEE Computer Society
Pages433-436
Number of pages4
ISBN (Print)9781479914814
DOIs
StatePublished - 2014
Event2014 IEEE 8th Sensor Array and Multichannel Signal Processing Workshop, SAM 2014 - A Coruna, Spain
Duration: Jun 22 2014Jun 25 2014

Publication series

NameProceedings of the IEEE Sensor Array and Multichannel Signal Processing Workshop
ISSN (Electronic)2151-870X

Other

Other2014 IEEE 8th Sensor Array and Multichannel Signal Processing Workshop, SAM 2014
CountrySpain
CityA Coruna
Period6/22/146/25/14

Fingerprint Dive into the research topics of 'Imputation of streaming low-rank tensor data'. Together they form a unique fingerprint.

Cite this