Toward Controlling Discrimination in Online Ad Auctions

Online advertising platforms are thriving due to the customizable audiences they offer advertisers. However, recent studies show that advertisements can be discriminatory with respect to the gender or race of the audience that sees the ad, and may inadvertently cross ethical and/or legal boundaries. To prevent this, we propose a constrained ad auction framework that maximizes the platform's revenue conditioned on ensuring that the audience seeing an advertiser's ad is distributed appropriately across sensitive types such as gender or race. Building upon Myerson's classic work, we first present an optimal auction mechanism for a large class of fairness constraints. Finding the parameters of this optimal auction, however, turns out to be a non-convex problem. We show that this non-convex problem can be reformulated as a more structured non-convex problem with no saddle points or local-maxima; this allows us to develop a gradient-descent-based algorithm to solve it. Our empirical results on the A1 Yahoo! dataset demonstrate that our algorithm can obtain uniform coverage across different user types for each advertiser at a minor loss to the revenue of the platform, and a small change to the size of the audience each advertiser reaches.

[1]  Nisheeth K. Vishnoi,et al.  Classification with Fairness Constraints: A Meta-Algorithm with Provable Guarantees , 2018, FAT.

[2]  Latanya Sweeney,et al.  Discrimination in Online Ad Delivery , 2013, ACM Queue.

[3]  Wen Zhang,et al.  How much can behavioral targeting help online advertising? , 2009, WWW '09.

[4]  Ayman Farahat How effective is targeted advertising? , 2013, ACC.

[5]  Michael Carl Tschantz,et al.  Automated Experiments on Ad Privacy Settings , 2014, Proc. Priv. Enhancing Technol..

[6]  William Vickrey,et al.  Counterspeculation, Auctions, And Competitive Sealed Tenders , 1961 .

[7]  Nisheeth K. Vishnoi,et al.  Controlling Polarization in Personalization: An Algorithmic Framework , 2019, FAT.

[8]  R. Myerson Incentive Compatibility and the Bargaining Problem , 1979 .

[9]  Gaël Varoquaux,et al.  Scikit-learn: Machine Learning in Python , 2011, J. Mach. Learn. Res..

[10]  H. Varian Online Ad Auctions , 2009 .

[11]  Debmalya Panigrahi,et al.  Online selection of diverse results , 2012, WSDM '12.

[12]  Jun Wang,et al.  Internet Advertising: An Interplay among Advertisers, Online Publishers, Ad Exchanges and Web Users , 2012, ArXiv.

[13]  Filip Radlinski,et al.  Learning diverse rankings with multi-armed bandits , 2008, ICML '08.

[14]  Krishna P. Gummadi,et al.  Potential for Discrimination in Online Targeted Advertising , 2018, FAT.

[15]  Avi Goldfarb,et al.  Online Display Advertising: Targeting and Obtrusiveness , 2011, Mark. Sci..

[16]  E. H. Clarke Multipart pricing of public goods , 1971 .

[17]  Theodore Groves,et al.  Incentives in Teams , 1973 .

[18]  Tim Roughgarden,et al.  Algorithmic Game Theory , 2007 .

[19]  Aristides Gionis,et al.  Balancing Opposing Views to Reduce Controversy , 2016, ArXiv.

[20]  Jason D. Hartline Mechanism Design and Approximation , 2014 .

[21]  Sergei Vassilvitskii,et al.  Bidding for Representative Allocations for Display Advertising , 2009, WINE.

[22]  Michael Ostrovsky,et al.  Reserve Prices in Internet Advertising Auctions: A Field Experiment , 2009, Journal of Political Economy.

[23]  Jon Feldman,et al.  Yield Optimization of Display Advertising with Ad Exchange , 2014, Manag. Sci..

[24]  R. Varga Geršgorin And His Circles , 2004 .

[25]  Catherine Tucker,et al.  Algorithmic bias? An empirical study into apparent gender-based discrimination in the display of STEM career ads , 2019 .

[26]  Nisheeth K. Vishnoi,et al.  A dashboard for controlling polarization in personalization , 2019, AI Commun..

[27]  Roger B. Myerson,et al.  Optimal Auction Design , 1981, Math. Oper. Res..

[28]  Mallesh M. Pai,et al.  Auction Design with Fairness Concerns: Subsidies vs. Set-Asides , 2012 .

[29]  Piotr Sapiezynski,et al.  Discrimination through Optimization , 2019, Proc. ACM Hum. Comput. Interact..

[30]  Nisheeth K. Vishnoi,et al.  Fair and Diverse DPP-based Data Summarization , 2018, ICML.

[31]  Nisheeth K. Vishnoi,et al.  Ranking with Fairness Constraints , 2017, ICALP.

[32]  Nisheeth K. Vishnoi,et al.  Multiwinner Voting with Fairness Constraints , 2017, IJCAI.

[33]  Abolfazl Asudeh,et al.  Designing Fair Ranking Schemes , 2017, SIGMOD Conference.

[34]  S. Muthukrishnan,et al.  Ad Exchanges: Research Issues , 2009, WINE.

[35]  Eric Jones,et al.  SciPy: Open Source Scientific Tools for Python , 2001 .

[36]  Krishna P. Gummadi,et al.  Fairness Constraints: Mechanisms for Fair Classification , 2015, AISTATS.