Hellinger distance based drift detection for nonstationary environments

Gregory Ditzler, Robi Polikar

    Research output: Chapter in Book/Report/Conference proceedingConference contribution

    57 Scopus citations

    Abstract

    Most machine learning algorithms, including many online learners, assume that the data distribution to be learned is fixed. There are many real-world problems where the distribution of the data changes as a function of time. Changes in nonstationary data distributions can significantly reduce the generalization ability of the learning algorithm on new or field data, if the algorithm is not equipped to track such changes. When the stationary data distribution assumption does not hold, the learner must take appropriate actions to ensure that the new/relevant information is learned. On the other hand, data distributions do not necessarily change continuously, necessitating the ability to monitor the distribution and detect when a significant change in distribution has occurred. In this work, we propose and analyze a feature based drift detection method using the Hellinger distance to detect gradual or abrupt changes in the distribution.

    Original languageEnglish (US)
    Title of host publicationIEEE SSCI 2011
    Subtitle of host publicationSymposium Series on Computational Intelligence - CIDUE 2011: 2011 IEEE Symposium on Computational Intelligence in Dynamic and Uncertain Environments
    Pages41-48
    Number of pages8
    DOIs
    StatePublished - 2011
    EventSymposium Series on Computational Intelligence, IEEE SSCI 2011 - 2011 IEEE Symposium on Computational Intelligence in Dynamic and Uncertain Environments, CIDUE 2011 - Paris, France
    Duration: Apr 11 2011Apr 15 2011

    Publication series

    NameIEEE SSCI 2011: Symposium Series on Computational Intelligence - CIDUE 2011: 2011 IEEE Symposium on Computational Intelligence in Dynamic and Uncertain Environments

    Other

    OtherSymposium Series on Computational Intelligence, IEEE SSCI 2011 - 2011 IEEE Symposium on Computational Intelligence in Dynamic and Uncertain Environments, CIDUE 2011
    Country/TerritoryFrance
    CityParis
    Period4/11/114/15/11

    All Science Journal Classification (ASJC) codes

    • Artificial Intelligence
    • Computational Theory and Mathematics
    • Computer Science Applications

    Fingerprint

    Dive into the research topics of 'Hellinger distance based drift detection for nonstationary environments'. Together they form a unique fingerprint.

    Cite this