Skip to main navigation Skip to search Skip to main content

Guarantees of augmented trace norm models in tensor recovery

  • Harbin Institute of Technology
  • Fujitsu
  • Beijing GuoDianTong Network Technology Co., Ltd.

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

Abstract

This paper studies the recovery guarantees of the models of minimizing ∥Χ∥* + 1/2α ∥Χ∥ F2 where Χ is a tensor and ∥Χ∥* and ∥Χ∥ F are the trace and Frobenius norm of respectively. We show that they can efficiently recover low-rank tensors. In particular, they enjoy exact guarantees similar to those known for minimizing ∥Χ∥* under the conditions on the sensing operator such as its null-space property, restricted isometry property, or spherical section property. To recover a low-rank tensor Χ0, minimizing ∥Χ∥* + 1/2α ∥Χ∥ F2 returns the same solution as minimizing ∥Χ∥* almost whenever α ≥ 10 imax ∥X(i)02.

Original languageEnglish
Title of host publicationIJCAI 2013 - Proceedings of the 23rd International Joint Conference on Artificial Intelligence
Pages1670-1676
Number of pages7
StatePublished - 2013
Event23rd International Joint Conference on Artificial Intelligence, IJCAI 2013 - Beijing, China
Duration: 3 Aug 20139 Aug 2013

Publication series

NameIJCAI International Joint Conference on Artificial Intelligence
ISSN (Print)1045-0823

Conference

Conference23rd International Joint Conference on Artificial Intelligence, IJCAI 2013
Country/TerritoryChina
CityBeijing
Period3/08/139/08/13

Fingerprint

Dive into the research topics of 'Guarantees of augmented trace norm models in tensor recovery'. Together they form a unique fingerprint.

Cite this