Learning Sparse Neural Networks for Better Generalization

    Onderzoeksoutput: Hoofdstuk in Boek/Rapport/CongresprocedureConferentiebijdrageAcademicpeer review

    61 Downloads (Pure)

    Samenvatting

    Deep neural networks perform well on test data when they are highly overparameterized, which, however, also leads to large cost to train and deploy them. As a leading approach to address this problem, sparse neural networks have been widely used to significantly reduce the size of networks, making them more efficient during training and deployment, without compromising performance. Recently, sparse neural networks, either compressed from a pre-trained model or obtained by training from scratch, have been observed to be able to generalize as well as or even better than their dense counterparts. However, conventional techniques to find well fitted sparse sub-networks are expensive and the mechanisms underlying this phenomenon are far from clear. To tackle these problems, this Ph.D. research aims to study the generalization of sparse neural networks, and to propose more efficient approaches that can yield sparse neural networks with generalization bounds.
    Originele taal-2Engels
    TitelInternational Joint Conferences on Artifical Intelligence (IJCAI), 2020
    Aantal pagina's2
    StatusGeaccepteerd/In druk - 20 apr 2020
    Evenement29th International Joint Conference on Artificial Intelligence - 17th Pacific Rim International Conference on Artificial Intelligence. - Pacifico Convention Plaza Yokohama, Yokohama, Japan
    Duur: 11 jul 202017 jul 2020
    Congresnummer: 29
    https://ijcai20.org/

    Congres

    Congres29th International Joint Conference on Artificial Intelligence - 17th Pacific Rim International Conference on Artificial Intelligence.
    Verkorte titelIJCAI-PRICAI 2020
    Land/RegioJapan
    StadYokohama
    Periode11/07/2017/07/20
    Internet adres

    Vingerafdruk

    Duik in de onderzoeksthema's van 'Learning Sparse Neural Networks for Better Generalization'. Samen vormen ze een unieke vingerafdruk.

    Citeer dit