Mixture augmented lagrange multiplier method for tensor recovery

Huachun Tan, Bin Cheng, Jianshuai Feng, Guangdong Feng, Wuhong Wang, Yu Jin Zhang

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

The problem of data recovery in multi-way arrays (i.e., tensors) arises in many fields such as image processing and computer vision, etc. In this paper we propose a scalable and fast algorithm for recovering a low-n-rank tensor with an unknown fraction of its entries being arbitrarily corrupted. In the new algorithm, the tensor recovery problem is formulated as a mixture convex multilinear RPCA optimization problem by minimizing a sum of the nuclear norm and the ℓ1-norm. The problem is well-structured in both the objective function and constraints. We apply augmented Lagrange multiplier method which can make use of the good structure for efficiently solving this problem.

Original languageEnglish
Title of host publicationWorld Scientific Proc. Series on Computer Engineering and Information Science 7; Uncertainty Modeling in Knowledge Engineering and Decision Making - Proceedings of the 10th International FLINS Conf.
PublisherWorld Scientific Publishing Co. Pte Ltd
Pages430-435
Number of pages6
ISBN (Print)9789814417730
DOIs
Publication statusPublished - 2012
Event10th International Fuzzy Logic and Intelligent Technologies inNuclear Science Conference, FLINS 2012 - Istanbul, Turkey
Duration: 26 Aug 201229 Aug 2012

Publication series

NameWorld Scientific Proc. Series on Computer Engineering and Information Science 7; Uncertainty Modeling in Knowledge Engineering and Decision Making - Proceedings of the 10th International FLINS Conf.
Volume7

Conference

Conference10th International Fuzzy Logic and Intelligent Technologies inNuclear Science Conference, FLINS 2012
Country/TerritoryTurkey
CityIstanbul
Period26/08/1229/08/12

Fingerprint

Dive into the research topics of 'Mixture augmented lagrange multiplier method for tensor recovery'. Together they form a unique fingerprint.

Cite this