Constrained Stochastic Gradient Descent: The Good Practice

Soumava Kumar Roy, Mehrtash Harandi

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

    7 Citations (Scopus)

    Abstract

    Stochastic Gradient Descent (SGD) is the method of choice for large scale problems, most notably in deep learning. Recent studies target improving convergence and speed of the SGD algorithm. In this paper, we equip the SGD algorithm and its advanced versions with an intriguing feature, namely handling constrained problems. Constraints such as orthogonality are pervasive in learning theory. Nevertheless and to some extent surprising, constrained SGD algorithms are rarely studied. Our proposal makes use of Riemannian geometry and accelerated optimization techniques to deliver efficient and constrained-aware SGD methods.We will assess and contrast our proposed approaches in a wide range of problems including incremental dimensionality reduction, karcher mean and deep metric learning.

    Original languageEnglish
    Title of host publicationDICTA 2017 - 2017 International Conference on Digital Image Computing
    Subtitle of host publicationTechniques and Applications
    EditorsYi Guo, Manzur Murshed, Zhiyong Wang, David Dagan Feng, Hongdong Li, Weidong Tom Cai, Junbin Gao
    PublisherInstitute of Electrical and Electronics Engineers Inc.
    Pages1-8
    Number of pages8
    ISBN (Electronic)9781538628393
    DOIs
    Publication statusPublished - 19 Dec 2017
    Event2017 International Conference on Digital Image Computing: Techniques and Applications, DICTA 2017 - Sydney, Australia
    Duration: 29 Nov 20171 Dec 2017

    Publication series

    NameDICTA 2017 - 2017 International Conference on Digital Image Computing: Techniques and Applications
    Volume2017-December

    Conference

    Conference2017 International Conference on Digital Image Computing: Techniques and Applications, DICTA 2017
    Country/TerritoryAustralia
    CitySydney
    Period29/11/171/12/17

    Fingerprint

    Dive into the research topics of 'Constrained Stochastic Gradient Descent: The Good Practice'. Together they form a unique fingerprint.

    Cite this