Statistical mechanics of support vector networks

Rainer Dietrich, Manfred Opper, Haim Sompolinsky

    Research output: Contribution to journalArticlepeer-review

    Abstract

    Using methods of Statistical Physics, we investigate the generalization performance of support vector machines (SVMs), which have been recently introduced as a general alternative to neural networks. For nonlinear classification rules, the generalization error saturates on a plateau, when the number of examples is too small to properly estimate the coefficients of the nonlinear part. When trained on simple rules, we find that SVMs overfit only weakly. The performance of SVMs is strongly enhanced, when the distribution of the inputs has a gap in feature space.
    Original languageEnglish
    Pages (from-to)2975-2978
    Number of pages4
    JournalPhysical Review Letters
    Volume82
    Issue number14
    DOIs
    Publication statusPublished - 5 Apr 1999

    Bibliographical note

    Copyright of the American Physical Society

    Keywords

    • statistical physics
    • support vector machines
    • neural networks
    • nonlinear classification
    • generalization error

    Fingerprint

    Dive into the research topics of 'Statistical mechanics of support vector networks'. Together they form a unique fingerprint.

    Cite this