• Login
    View Item 
    •   Eurographics DL Home
    • Computer Graphics Forum
    • Volume 40 (2021)
    • 40-Issue 1
    • View Item
    •   Eurographics DL Home
    • Computer Graphics Forum
    • Volume 40 (2021)
    • 40-Issue 1
    • View Item
    JavaScript is disabled for your browser. Some features of this site may not work without it.

    TopoAct: Visually Exploring the Shape of Activations in Deep Learning

    Thumbnail
    View/Open
    v40i1pp382-397.pdf (5.928Mb)
    Date
    2021
    Author
    Rathore, Archit
    Chalapathi, Nithin
    Palande, Sourabh
    Wang, Bei
    Pay-Per-View via TIB Hannover:

    Try if this item/paper is available.

    Metadata
    Show full item record
    Abstract
    Deep neural networks such as GoogLeNet, ResNet, and BERT have achieved impressive performance in tasks such as image and text classification. To understand how such performance is achieved, we probe a trained deep neural network by studying neuron activations, i.e.combinations of neuron firings, at various layers of the network in response to a particular input. With a large number of inputs, we aim to obtain a global view of what neurons detect by studying their activations. In particular, we develop visualizations that show the shape of the activation space, the organizational principle behind neuron activations, and the relationships of these activations within a layer. Applying tools from topological data analysis, we present , a visual exploration system to study topological summaries of activation vectors. We present exploration scenarios using that provide valuable insights into learned representations of neural networks. We expect to give a topological perspective that enriches the current toolbox of neural network analysis, and to provide a basis for network architecture diagnosis and data anomaly detection.
    BibTeX
    @article {10.1111:cgf.14195,
    journal = {Computer Graphics Forum},
    title = {{TopoAct: Visually Exploring the Shape of Activations in Deep Learning}},
    author = {Rathore, Archit and Chalapathi, Nithin and Palande, Sourabh and Wang, Bei},
    year = {2021},
    publisher = {© 2021 Eurographics ‐ The European Association for Computer Graphics and John Wiley & Sons Ltd},
    ISSN = {1467-8659},
    DOI = {10.1111/cgf.14195}
    }
    URI
    https://doi.org/10.1111/cgf.14195
    https://diglib.eg.org:443/handle/10.1111/cgf14195
    Collections
    • 40-Issue 1

    Eurographics Association copyright © 2013 - 2023 
    Send Feedback | Contact - Imprint | Data Privacy Policy | Disable Google Analytics
    Theme by @mire NV
    System hosted at  Graz University of Technology.
    TUGFhA
     

     

    Browse

    All of Eurographics DLCommunities & CollectionsBy Issue DateAuthorsTitlesSubjectsThis CollectionBy Issue DateAuthorsTitlesSubjects

    My Account

    LoginRegister

    Statistics

    View Usage Statistics

    BibTeX | TOC

    Create BibTeX Create Table of Contents

    Eurographics Association copyright © 2013 - 2023 
    Send Feedback | Contact - Imprint | Data Privacy Policy | Disable Google Analytics
    Theme by @mire NV
    System hosted at  Graz University of Technology.
    TUGFhA