• Login
    JavaScript is disabled for your browser. Some features of this site may not work without it.
    The Effects of Noisy Labels on Deep Convolutional Neural Networks for Music Tagging 
    •   QMRO Home
    • School of Electronic Engineering and Computer Science
    • Electronic Engineering and Computer Science
    • The Effects of Noisy Labels on Deep Convolutional Neural Networks for Music Tagging
    •   QMRO Home
    • School of Electronic Engineering and Computer Science
    • Electronic Engineering and Computer Science
    • The Effects of Noisy Labels on Deep Convolutional Neural Networks for Music Tagging
    ‌
    ‌

    Browse

    All of QMROCommunities & CollectionsBy Issue DateAuthorsTitlesSubjectsThis CollectionBy Issue DateAuthorsTitlesSubjects
    ‌
    ‌

    Administrators only

    Login
    ‌
    ‌

    Statistics

    Most Popular ItemsStatistics by CountryMost Popular Authors

    The Effects of Noisy Labels on Deep Convolutional Neural Networks for Music Tagging

    View/Open
    Accepted Version (2.702Mb)
    Volume
    2
    Pagination
    139 - 149
    Publisher
    IEEE
    Publisher URL
    https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=8323324
    DOI
    10.1109/TETCI.2017.2771298
    Journal
    IEEE Transactions on Emerging Topics in Computational Intelligence
    Issue
    2
    Metadata
    Show full item record
    Abstract
    Deep neural networks (DNN) have been successfully applied to music classification including music tagging. However, there are several open questions regarding the training, evaluation, and analysis of DNNs. In this article, we investigate specific aspects of neural networks, the effects of noisy labels, to deepen our understanding of their properties. We analyse and (re-)validate a large music tagging dataset to investigate the reliability of training and evaluation. Using a trained network, we compute label vector similarities which is compared to groundtruth similarity. The results highlight several important aspects of music tagging and neural networks. We show that networks can be effective despite relatively large error rates in groundtruth datasets, while conjecturing that label noise can be the cause of varying tag-wise performance differences. Lastly, the analysis of our trained network provides valuable insight into the relationships between music tags. These results highlight the benefit of using data-driven methods to address automatic music tagging.
    Authors
    Choi, K; Fazekas, G; Sandler, M; Cho, K
    URI
    http://qmro.qmul.ac.uk/xmlui/handle/123456789/30523
    Collections
    • Electronic Engineering and Computer Science [2959]
    Language
    English
    Copyright statements
    © 2017 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, 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 component of this work in other works.
    Twitter iconFollow QMUL on Twitter
    Twitter iconFollow QM Research
    Online on twitter
    Facebook iconLike us on Facebook
    • Site Map
    • Privacy and cookies
    • Disclaimer
    • Accessibility
    • Contacts
    • Intranet
    • Current students

    Modern Slavery Statement

    Queen Mary University of London
    Mile End Road
    London E1 4NS
    Tel: +44 (0)20 7882 5555

    © Queen Mary University of London.