BIROn - Birkbeck Institutional Research Online

    Self-taught semi-supervised dictionary learning with non-negative constraint

    Zhang, X. and Liu, Q. and Wang, D. and Zhao, L. and Gu, N. and Maybank, Stephen J. (2019) Self-taught semi-supervised dictionary learning with non-negative constraint. IEEE Transactions on Industrrial Informatics 16 (1), pp. 532-543. ISSN 1551-3203.

    SelfTaught.pdf - Author's Accepted Manuscript

    Download (1MB) | Preview


    This paper investigates classification by dictionary learning. A novel unified framework termed self-taught semisupervised dictionary learning with non-negative constraint (NNST-SSDL) is proposed for simultaneously optimizing the components of a dictionary and a graph Laplacian. Specifically, an atom graph Laplacian regularization is built by using sparse coefficients to effectively capture the underlying manifold structure. It is more robust to noisy samples and outliers because atoms are more concise and representative than training samples. A non-negative constraint imposed on the sparse coefficients guarantees that each sample is in the middle of its related atoms. In this way the dependency between samples and atoms is made explicit. Furthermore, a self-taught mechanism is introduced to effectively feed back the manifold structure induced by atom graph Laplacian regularization and the supervised information hidden in unlabeled samples in order to learn a better dictionary. An efficient algorithm, combining a block coordinate descent method with the alternating direction method of multipliers is derived to optimize the unified framework. Experimental results on several benchmark datasets show the effectiveness of the proposed model.


    Item Type: Article
    Additional Information: (c) 2019 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other users, including reprinting/ republishing this material for advertising or promotional purposes, creating new collective works for resale or redistribution to servers or lists, or reuse of any copyrighted components of this work in other works.
    Keyword(s) / Subject(s): Semi-supervised dictionary learning, nonnegative constraint, atom graph regularization, self-taught
    School: School of Business, Economics & Informatics > Computer Science and Information Systems
    Depositing User: Stephen Maybank
    Date Deposited: 23 Jul 2019 12:33
    Last Modified: 06 Sep 2022 10:06


    Activity Overview
    6 month trend
    6 month trend

    Additional statistics are available via IRStats2.

    Archive Staff Only (login required)

    Edit/View Item Edit/View Item