Revisiting the approximation bound for stochastic submodular cover

Lisa Hellerstein, Devorah Kletenik

    Research output: Contribution to journalArticle

    Abstract

    Deshpande et al. presented a k(lnR+1) approximation bound for Stochastic Submodular Cover, where k is the state set size, R is the maximum utility of a single item, and the utility function is integer-valued. This bound is similar to the (ln Q= + 1) bound given by Golovin and Krause, whose analysis was recently found to have an error. Here Q ≥ R is the goal utility and Q is the minimum gap between Q and any attainable utility Q0 < Q. We revisit the proof of the k(lnR+1) bound of Deshpande et al., fill in the details of the proof of a key lemma, and prove two bounds for real-valued utility functions: k(lnR1 +1) and (lnRE +1). Here R1 equals the maximum ratio between the largest increase in utility attainable from a single item, and the smallest non-zero increase attainable from that same item (in the same state). The quantity RE equals the maximum ratio between the largest expected increase in utility from a single item, and the smallest non-zero expected increase in utility from that same item. Our bounds apply only to the stochastic setting with independent states.

    Original languageEnglish (US)
    Pages (from-to)265-279
    Number of pages15
    JournalJournal of Artificial Intelligence Research
    Volume63
    StatePublished - Oct 1 2018

    ASJC Scopus subject areas

    • Artificial Intelligence

    Cite this

    Revisiting the approximation bound for stochastic submodular cover. / Hellerstein, Lisa; Kletenik, Devorah.

    In: Journal of Artificial Intelligence Research, Vol. 63, 01.10.2018, p. 265-279.

    Research output: Contribution to journalArticle

    Hellerstein, Lisa ; Kletenik, Devorah. / Revisiting the approximation bound for stochastic submodular cover. In: Journal of Artificial Intelligence Research. 2018 ; Vol. 63. pp. 265-279.
    @article{2d2f5040c9ae4d93bbaaa75021929d7c,
    title = "Revisiting the approximation bound for stochastic submodular cover",
    abstract = "Deshpande et al. presented a k(lnR+1) approximation bound for Stochastic Submodular Cover, where k is the state set size, R is the maximum utility of a single item, and the utility function is integer-valued. This bound is similar to the (ln Q= + 1) bound given by Golovin and Krause, whose analysis was recently found to have an error. Here Q ≥ R is the goal utility and Q is the minimum gap between Q and any attainable utility Q0 < Q. We revisit the proof of the k(lnR+1) bound of Deshpande et al., fill in the details of the proof of a key lemma, and prove two bounds for real-valued utility functions: k(lnR1 +1) and (lnRE +1). Here R1 equals the maximum ratio between the largest increase in utility attainable from a single item, and the smallest non-zero increase attainable from that same item (in the same state). The quantity RE equals the maximum ratio between the largest expected increase in utility from a single item, and the smallest non-zero expected increase in utility from that same item. Our bounds apply only to the stochastic setting with independent states.",
    author = "Lisa Hellerstein and Devorah Kletenik",
    year = "2018",
    month = "10",
    day = "1",
    language = "English (US)",
    volume = "63",
    pages = "265--279",
    journal = "Journal of Artificial Intelligence Research",
    issn = "1076-9757",
    publisher = "Morgan Kaufmann Publishers, Inc.",

    }

    TY - JOUR

    T1 - Revisiting the approximation bound for stochastic submodular cover

    AU - Hellerstein, Lisa

    AU - Kletenik, Devorah

    PY - 2018/10/1

    Y1 - 2018/10/1

    N2 - Deshpande et al. presented a k(lnR+1) approximation bound for Stochastic Submodular Cover, where k is the state set size, R is the maximum utility of a single item, and the utility function is integer-valued. This bound is similar to the (ln Q= + 1) bound given by Golovin and Krause, whose analysis was recently found to have an error. Here Q ≥ R is the goal utility and Q is the minimum gap between Q and any attainable utility Q0 < Q. We revisit the proof of the k(lnR+1) bound of Deshpande et al., fill in the details of the proof of a key lemma, and prove two bounds for real-valued utility functions: k(lnR1 +1) and (lnRE +1). Here R1 equals the maximum ratio between the largest increase in utility attainable from a single item, and the smallest non-zero increase attainable from that same item (in the same state). The quantity RE equals the maximum ratio between the largest expected increase in utility from a single item, and the smallest non-zero expected increase in utility from that same item. Our bounds apply only to the stochastic setting with independent states.

    AB - Deshpande et al. presented a k(lnR+1) approximation bound for Stochastic Submodular Cover, where k is the state set size, R is the maximum utility of a single item, and the utility function is integer-valued. This bound is similar to the (ln Q= + 1) bound given by Golovin and Krause, whose analysis was recently found to have an error. Here Q ≥ R is the goal utility and Q is the minimum gap between Q and any attainable utility Q0 < Q. We revisit the proof of the k(lnR+1) bound of Deshpande et al., fill in the details of the proof of a key lemma, and prove two bounds for real-valued utility functions: k(lnR1 +1) and (lnRE +1). Here R1 equals the maximum ratio between the largest increase in utility attainable from a single item, and the smallest non-zero increase attainable from that same item (in the same state). The quantity RE equals the maximum ratio between the largest expected increase in utility from a single item, and the smallest non-zero expected increase in utility from that same item. Our bounds apply only to the stochastic setting with independent states.

    UR - http://www.scopus.com/inward/record.url?scp=85055880835&partnerID=8YFLogxK

    UR - http://www.scopus.com/inward/citedby.url?scp=85055880835&partnerID=8YFLogxK

    M3 - Article

    AN - SCOPUS:85055880835

    VL - 63

    SP - 265

    EP - 279

    JO - Journal of Artificial Intelligence Research

    JF - Journal of Artificial Intelligence Research

    SN - 1076-9757

    ER -