Learning in the presence of finitely or infinitely many irrelevant attributes

Avrim Blum, Lisa Hellerstein, Nick Littlestone

    Research output: Contribution to journalArticle

    Abstract

    The problem of learning boolean functions in query and mistake-bound models in the presence of irrelevant attributes is addressed. Emphasis is directed towards the learnability of the concept classes and whether the classes can be learned by an algorithm that is attribute-efficient. Results obtained are discussed in terms of the projection and embedding-closed (p.e.c) concept classes.

    Original languageEnglish (US)
    Pages (from-to)32-40
    Number of pages9
    JournalJournal of Computer and System Sciences
    Volume50
    Issue number1
    DOIs
    StatePublished - Feb 1995

    Fingerprint

    Boolean functions
    Attribute
    Learnability
    Boolean Functions
    Projection
    Query
    Closed
    Learning
    Class
    Concepts
    Model

    ASJC Scopus subject areas

    • Computational Theory and Mathematics
    • Theoretical Computer Science
    • Applied Mathematics
    • Computer Networks and Communications

    Cite this

    Learning in the presence of finitely or infinitely many irrelevant attributes. / Blum, Avrim; Hellerstein, Lisa; Littlestone, Nick.

    In: Journal of Computer and System Sciences, Vol. 50, No. 1, 02.1995, p. 32-40.

    Research output: Contribution to journalArticle

    Blum, Avrim ; Hellerstein, Lisa ; Littlestone, Nick. / Learning in the presence of finitely or infinitely many irrelevant attributes. In: Journal of Computer and System Sciences. 1995 ; Vol. 50, No. 1. pp. 32-40.
    @article{d1283d71aacc4de68ea33bb23494e1f3,
    title = "Learning in the presence of finitely or infinitely many irrelevant attributes",
    abstract = "The problem of learning boolean functions in query and mistake-bound models in the presence of irrelevant attributes is addressed. Emphasis is directed towards the learnability of the concept classes and whether the classes can be learned by an algorithm that is attribute-efficient. Results obtained are discussed in terms of the projection and embedding-closed (p.e.c) concept classes.",
    author = "Avrim Blum and Lisa Hellerstein and Nick Littlestone",
    year = "1995",
    month = "2",
    doi = "10.1006/jcss.1995.1004",
    language = "English (US)",
    volume = "50",
    pages = "32--40",
    journal = "Journal of Computer and System Sciences",
    issn = "0022-0000",
    publisher = "Academic Press Inc.",
    number = "1",

    }

    TY - JOUR

    T1 - Learning in the presence of finitely or infinitely many irrelevant attributes

    AU - Blum, Avrim

    AU - Hellerstein, Lisa

    AU - Littlestone, Nick

    PY - 1995/2

    Y1 - 1995/2

    N2 - The problem of learning boolean functions in query and mistake-bound models in the presence of irrelevant attributes is addressed. Emphasis is directed towards the learnability of the concept classes and whether the classes can be learned by an algorithm that is attribute-efficient. Results obtained are discussed in terms of the projection and embedding-closed (p.e.c) concept classes.

    AB - The problem of learning boolean functions in query and mistake-bound models in the presence of irrelevant attributes is addressed. Emphasis is directed towards the learnability of the concept classes and whether the classes can be learned by an algorithm that is attribute-efficient. Results obtained are discussed in terms of the projection and embedding-closed (p.e.c) concept classes.

    UR - http://www.scopus.com/inward/record.url?scp=0029254047&partnerID=8YFLogxK

    UR - http://www.scopus.com/inward/citedby.url?scp=0029254047&partnerID=8YFLogxK

    U2 - 10.1006/jcss.1995.1004

    DO - 10.1006/jcss.1995.1004

    M3 - Article

    VL - 50

    SP - 32

    EP - 40

    JO - Journal of Computer and System Sciences

    JF - Journal of Computer and System Sciences

    SN - 0022-0000

    IS - 1

    ER -