Sponsored search auctions with markovian users

Gagan Aggarwal, Jon Feldman, Shanmugavelayutham Muthukrishnan, Martin Pál

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

    Abstract

    Sponsored search involves running an auction among advertisers who bid in order to have their ad shown next to search results for specific keywords. The most popular auction for sponsored search is the "Generalized Second Price" (GSP) auction where advertisers are assigned to slots in the decreasing order of their score, which is defined as the product of their bid and click-through rate. One of the main advantages of this simple ranking is that bidding strategy is intuitive: to move up to a more prominent slot on the results page, bid more. This makes it simple for advertisers to strategize. However this ranking only maximizes efficiency under the assumption that the probability of a user clicking on an ad is independent of the other ads shown on the page. We study a Markovian user model that does not make this assumption. Under this model, the most efficient assignment is no longer a simple ranking function as in GSP. We show that the optimal assignment can be found efficiently (even in near-linear time). As a result of the more sophisticated structure of the optimal assignment, bidding dynamics become more complex: indeed it is no longer clear that bidding more moves one higher on the page. Our main technical result is that despite the added complexity of the bidding dynamics, the optimal assignment has the property that ad position is still monotone in bid. Thus even in this richer user model, our mechanism retains the core bidding dynamics of the GSP auction that make it useful for advertisers.

    Original languageEnglish (US)
    Title of host publicationInternet and Network Economics - 4th International Workshop, WINE 2008, Proceedings
    Pages621-628
    Number of pages8
    DOIs
    StatePublished - Dec 1 2008
    Event4th International Workshop on Internet and Network Economics, WINE 2008 - Shanghai, China
    Duration: Dec 17 2008Dec 20 2008

    Publication series

    NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
    Volume5385 LNCS
    ISSN (Print)0302-9743
    ISSN (Electronic)1611-3349

    Conference

    Conference4th International Workshop on Internet and Network Economics, WINE 2008
    CountryChina
    CityShanghai
    Period12/17/0812/20/08

    Fingerprint

    Bidding
    Auctions
    Assignment
    User Model
    Ranking
    Ranking Function
    Linear Time
    Intuitive
    Monotone
    Maximise

    ASJC Scopus subject areas

    • Theoretical Computer Science
    • Computer Science(all)

    Cite this

    Aggarwal, G., Feldman, J., Muthukrishnan, S., & Pál, M. (2008). Sponsored search auctions with markovian users. In Internet and Network Economics - 4th International Workshop, WINE 2008, Proceedings (pp. 621-628). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 5385 LNCS). https://doi.org/10.1007/978-3-540-92185-1_68

    Sponsored search auctions with markovian users. / Aggarwal, Gagan; Feldman, Jon; Muthukrishnan, Shanmugavelayutham; Pál, Martin.

    Internet and Network Economics - 4th International Workshop, WINE 2008, Proceedings. 2008. p. 621-628 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 5385 LNCS).

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

    Aggarwal, G, Feldman, J, Muthukrishnan, S & Pál, M 2008, Sponsored search auctions with markovian users. in Internet and Network Economics - 4th International Workshop, WINE 2008, Proceedings. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 5385 LNCS, pp. 621-628, 4th International Workshop on Internet and Network Economics, WINE 2008, Shanghai, China, 12/17/08. https://doi.org/10.1007/978-3-540-92185-1_68
    Aggarwal G, Feldman J, Muthukrishnan S, Pál M. Sponsored search auctions with markovian users. In Internet and Network Economics - 4th International Workshop, WINE 2008, Proceedings. 2008. p. 621-628. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). https://doi.org/10.1007/978-3-540-92185-1_68
    Aggarwal, Gagan ; Feldman, Jon ; Muthukrishnan, Shanmugavelayutham ; Pál, Martin. / Sponsored search auctions with markovian users. Internet and Network Economics - 4th International Workshop, WINE 2008, Proceedings. 2008. pp. 621-628 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).
    @inproceedings{a9ba428e159c4f85a08a1230cac82f0b,
    title = "Sponsored search auctions with markovian users",
    abstract = "Sponsored search involves running an auction among advertisers who bid in order to have their ad shown next to search results for specific keywords. The most popular auction for sponsored search is the {"}Generalized Second Price{"} (GSP) auction where advertisers are assigned to slots in the decreasing order of their score, which is defined as the product of their bid and click-through rate. One of the main advantages of this simple ranking is that bidding strategy is intuitive: to move up to a more prominent slot on the results page, bid more. This makes it simple for advertisers to strategize. However this ranking only maximizes efficiency under the assumption that the probability of a user clicking on an ad is independent of the other ads shown on the page. We study a Markovian user model that does not make this assumption. Under this model, the most efficient assignment is no longer a simple ranking function as in GSP. We show that the optimal assignment can be found efficiently (even in near-linear time). As a result of the more sophisticated structure of the optimal assignment, bidding dynamics become more complex: indeed it is no longer clear that bidding more moves one higher on the page. Our main technical result is that despite the added complexity of the bidding dynamics, the optimal assignment has the property that ad position is still monotone in bid. Thus even in this richer user model, our mechanism retains the core bidding dynamics of the GSP auction that make it useful for advertisers.",
    author = "Gagan Aggarwal and Jon Feldman and Shanmugavelayutham Muthukrishnan and Martin P{\'a}l",
    year = "2008",
    month = "12",
    day = "1",
    doi = "10.1007/978-3-540-92185-1_68",
    language = "English (US)",
    isbn = "3540921842",
    series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
    pages = "621--628",
    booktitle = "Internet and Network Economics - 4th International Workshop, WINE 2008, Proceedings",

    }

    TY - GEN

    T1 - Sponsored search auctions with markovian users

    AU - Aggarwal, Gagan

    AU - Feldman, Jon

    AU - Muthukrishnan, Shanmugavelayutham

    AU - Pál, Martin

    PY - 2008/12/1

    Y1 - 2008/12/1

    N2 - Sponsored search involves running an auction among advertisers who bid in order to have their ad shown next to search results for specific keywords. The most popular auction for sponsored search is the "Generalized Second Price" (GSP) auction where advertisers are assigned to slots in the decreasing order of their score, which is defined as the product of their bid and click-through rate. One of the main advantages of this simple ranking is that bidding strategy is intuitive: to move up to a more prominent slot on the results page, bid more. This makes it simple for advertisers to strategize. However this ranking only maximizes efficiency under the assumption that the probability of a user clicking on an ad is independent of the other ads shown on the page. We study a Markovian user model that does not make this assumption. Under this model, the most efficient assignment is no longer a simple ranking function as in GSP. We show that the optimal assignment can be found efficiently (even in near-linear time). As a result of the more sophisticated structure of the optimal assignment, bidding dynamics become more complex: indeed it is no longer clear that bidding more moves one higher on the page. Our main technical result is that despite the added complexity of the bidding dynamics, the optimal assignment has the property that ad position is still monotone in bid. Thus even in this richer user model, our mechanism retains the core bidding dynamics of the GSP auction that make it useful for advertisers.

    AB - Sponsored search involves running an auction among advertisers who bid in order to have their ad shown next to search results for specific keywords. The most popular auction for sponsored search is the "Generalized Second Price" (GSP) auction where advertisers are assigned to slots in the decreasing order of their score, which is defined as the product of their bid and click-through rate. One of the main advantages of this simple ranking is that bidding strategy is intuitive: to move up to a more prominent slot on the results page, bid more. This makes it simple for advertisers to strategize. However this ranking only maximizes efficiency under the assumption that the probability of a user clicking on an ad is independent of the other ads shown on the page. We study a Markovian user model that does not make this assumption. Under this model, the most efficient assignment is no longer a simple ranking function as in GSP. We show that the optimal assignment can be found efficiently (even in near-linear time). As a result of the more sophisticated structure of the optimal assignment, bidding dynamics become more complex: indeed it is no longer clear that bidding more moves one higher on the page. Our main technical result is that despite the added complexity of the bidding dynamics, the optimal assignment has the property that ad position is still monotone in bid. Thus even in this richer user model, our mechanism retains the core bidding dynamics of the GSP auction that make it useful for advertisers.

    UR - http://www.scopus.com/inward/record.url?scp=58849157872&partnerID=8YFLogxK

    UR - http://www.scopus.com/inward/citedby.url?scp=58849157872&partnerID=8YFLogxK

    U2 - 10.1007/978-3-540-92185-1_68

    DO - 10.1007/978-3-540-92185-1_68

    M3 - Conference contribution

    AN - SCOPUS:58849157872

    SN - 3540921842

    SN - 9783540921844

    T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

    SP - 621

    EP - 628

    BT - Internet and Network Economics - 4th International Workshop, WINE 2008, Proceedings

    ER -