Please use this identifier to cite or link to this item:
Scopus Web of Science® Altmetric
Full metadata record
DC FieldValueLanguage
dc.contributor.authorShen, C.-
dc.contributor.authorKim, J.-
dc.contributor.authorLiu, F.-
dc.contributor.authorWang, L.-
dc.contributor.authorvan den Hengel, A.-
dc.identifier.citationIEEE Transactions on Neural Networks and Learning Systems, 2014; 25(2):394-406-
dc.description.abstractDistance metric learning is of fundamental interest in machine learning because the employed distance metric can significantly affect the performance of many learning methods. Quadratic Mahalanobis metric learning is a popular approach to the problem, but typically requires solving a semidefinite programming (SDP) problem, which is computationally expensive. The worst case complexity of solving an SDP problem involving a matrix variable of size D × D with O(D) linear constraints is about O(D6.5) using interior-point methods, where D is the dimension of the input data. Thus, the interior-point methods only practically solve problems exhibiting less than a few thousand variables. Because the number of variables is D(D + 1)/2, this implies a limit upon the size of problem that can practically be solved around a few hundred dimensions. The complexity of the popular quadratic Mahalanobis metric learning approach thus limits the size of problem to which metric learning can be applied. Here, we propose a significantly more efficient and scalable approach to the metric learning problem based on the Lagrange dual formulation of the problem. The proposed formulation is much simpler to implement, and therefore allows much larger Mahalanobis metric learning problems to be solved. The time complexity of the proposed method is roughly O(D3), which is significantly lower than that of the SDP approach. Experiments on a variety of data sets demonstrate that the proposed method achieves an accuracy comparable with the state of the art, but is applicable to significantly larger problems. We also show that the proposed method can be applied to solve more general Frobenius norm regularized SDP problems approximately.-
dc.description.statementofresponsibilityChunhua Shen, Junae Kim, Fayao Liu, Lei Wang and Anton van den Hengel-
dc.publisherInstitute of Electrical and Electronics Engineers-
dc.rights© 2013 IEEE-
dc.subjectConvex optimization; Lagrange duality; Mahalanobis distance; metric learning; semidefinite programming (SDP)-
dc.titleEfficient dual approach to distance metric learning-
dc.typeJournal article-
dc.identifier.orcidShen, C. [0000-0002-8648-8718]-
dc.identifier.orcidvan den Hengel, A. [0000-0003-3027-8364]-
Appears in Collections:Aurora harvest 7
Computer Science publications

Files in This Item:
There are no files associated with this item.

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.