View in EDS

LSTMVis: A Tool for Visual Analysis of Hidden State Dynamics in Recurrent Neural Networks.

Saved in:
Bibliographic Details
Title: LSTMVis: A Tool for Visual Analysis of Hidden State Dynamics in Recurrent Neural Networks.
Authors: Strobelt, Hendrik1, Gehrmann, Sebastian1, Pfister, Hanspeter1, Rush, Alexander M.1
Source: IEEE Transactions on Visualization & Computer Graphics. Jan2018, Vol. 24 Issue 1, p667-676. 10p.
Subjects: Recurrent neural networks, Visual analytics, Data visualization, Big data, Machine learning, Computer software, Management
Abstract: Recurrent neural networks, and in particular long short-term memory (LSTM) networks, are a remarkably effective tool for sequence modeling that learn a dense black-box hidden representation of their sequential input. Researchers interested in better understanding these models have studied the changes in hidden state representations over time and noticed some interpretable patterns but also significant noise. In this work, we present LSTMVis, a visual analysis tool for recurrent neural networks with a focus on understanding these hidden state dynamics. The tool allows users to select a hypothesis input range to focus on local state changes, to match these states changes to similar patterns in a large data set, and to align these results with structural annotations from their domain. We show several use cases of the tool for analyzing specific hidden state properties on dataset containing nesting, phrase structure, and chord progressions, and demonstrate how the tool can be used to isolate patterns for further statistical analysis. We characterize the domain, the different stakeholders, and their goals and tasks. Long-term usage data after putting the tool online revealed great interest in the machine learning community. [ABSTRACT FROM PUBLISHER]
Copyright of IEEE Transactions on Visualization & Computer Graphics is the property of IEEE and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)
Database: Engineering Source
FullText Text:
  Availability: 0
Header DbId: egs
DbLabel: Engineering Source
An: 126653845
AccessLevel: 6
PubType: Academic Journal
PubTypeId: academicJournal
PreciseRelevancyScore: 0
IllustrationInfo
Items – Name: Title
  Label: Title
  Group: Ti
  Data: LSTMVis: A Tool for Visual Analysis of Hidden State Dynamics in Recurrent Neural Networks.
– Name: Author
  Label: Authors
  Group: Au
  Data: <searchLink fieldCode="AR" term="%22Strobelt%2C+Hendrik%22">Strobelt, Hendrik</searchLink><relatesTo>1</relatesTo><br /><searchLink fieldCode="AR" term="%22Gehrmann%2C+Sebastian%22">Gehrmann, Sebastian</searchLink><relatesTo>1</relatesTo><br /><searchLink fieldCode="AR" term="%22Pfister%2C+Hanspeter%22">Pfister, Hanspeter</searchLink><relatesTo>1</relatesTo><br /><searchLink fieldCode="AR" term="%22Rush%2C+Alexander+M%2E%22">Rush, Alexander M.</searchLink><relatesTo>1</relatesTo>
– Name: TitleSource
  Label: Source
  Group: Src
  Data: <searchLink fieldCode="JN" term="%22IEEE+Transactions+on+Visualization+%26+Computer+Graphics%22">IEEE Transactions on Visualization & Computer Graphics</searchLink>. Jan2018, Vol. 24 Issue 1, p667-676. 10p.
– Name: Subject
  Label: Subjects
  Group: Su
  Data: <searchLink fieldCode="DE" term="%22Recurrent+neural+networks%22">Recurrent neural networks</searchLink><br /><searchLink fieldCode="DE" term="%22Visual+analytics%22">Visual analytics</searchLink><br /><searchLink fieldCode="DE" term="%22Data+visualization%22">Data visualization</searchLink><br /><searchLink fieldCode="DE" term="%22Big+data%22">Big data</searchLink><br /><searchLink fieldCode="DE" term="%22Machine+learning%22">Machine learning</searchLink><br /><searchLink fieldCode="DE" term="%22Computer+software%22">Computer software</searchLink><br /><searchLink fieldCode="DE" term="%22Management%22">Management</searchLink>
– Name: Abstract
  Label: Abstract
  Group: Ab
  Data: Recurrent neural networks, and in particular long short-term memory (LSTM) networks, are a remarkably effective tool for sequence modeling that learn a dense black-box hidden representation of their sequential input. Researchers interested in better understanding these models have studied the changes in hidden state representations over time and noticed some interpretable patterns but also significant noise. In this work, we present LSTMVis, a visual analysis tool for recurrent neural networks with a focus on understanding these hidden state dynamics. The tool allows users to select a hypothesis input range to focus on local state changes, to match these states changes to similar patterns in a large data set, and to align these results with structural annotations from their domain. We show several use cases of the tool for analyzing specific hidden state properties on dataset containing nesting, phrase structure, and chord progressions, and demonstrate how the tool can be used to isolate patterns for further statistical analysis. We characterize the domain, the different stakeholders, and their goals and tasks. Long-term usage data after putting the tool online revealed great interest in the machine learning community. [ABSTRACT FROM PUBLISHER]
– Name: AbstractSuppliedCopyright
  Label:
  Group: Ab
  Data: <i>Copyright of IEEE Transactions on Visualization & Computer Graphics is the property of IEEE and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract.</i> (Copyright applies to all Abstracts.)
PLink https://search.ebscohost.com/login.aspx?direct=true&site=eds-live&db=egs&AN=126653845
RecordInfo BibRecord:
  BibEntity:
    Identifiers:
      – Type: doi
        Value: 10.1109/TVCG.2017.2744158
    Languages:
      – Code: eng
        Text: English
    PhysicalDescription:
      Pagination:
        PageCount: 10
        StartPage: 667
    Subjects:
      – SubjectFull: Recurrent neural networks
        Type: general
      – SubjectFull: Visual analytics
        Type: general
      – SubjectFull: Data visualization
        Type: general
      – SubjectFull: Big data
        Type: general
      – SubjectFull: Machine learning
        Type: general
      – SubjectFull: Computer software
        Type: general
      – SubjectFull: Management
        Type: general
    Titles:
      – TitleFull: LSTMVis: A Tool for Visual Analysis of Hidden State Dynamics in Recurrent Neural Networks.
        Type: main
  BibRelationships:
    HasContributorRelationships:
      – PersonEntity:
          Name:
            NameFull: Strobelt, Hendrik
      – PersonEntity:
          Name:
            NameFull: Gehrmann, Sebastian
      – PersonEntity:
          Name:
            NameFull: Pfister, Hanspeter
      – PersonEntity:
          Name:
            NameFull: Rush, Alexander M.
    IsPartOfRelationships:
      – BibEntity:
          Dates:
            – D: 01
              M: 01
              Text: Jan2018
              Type: published
              Y: 2018
          Identifiers:
            – Type: issn-print
              Value: 10772626
          Numbering:
            – Type: volume
              Value: 24
            – Type: issue
              Value: 1
          Titles:
            – TitleFull: IEEE Transactions on Visualization & Computer Graphics
              Type: main
ResultId 1