ABSTRACT
In a typical learning problem, one key step is to use training data to pick one model from a collection of models that optimizes an objective function. In many multi-agent settings, the training data is generated through the actions of the agents, and the model is used to make a decision (e.g., how to sell an item) that affects the agents. An illustrative example of this is the problem of learning the reserve price in an auction. In such cases, the agents have an incentive to influence the training data (e.g., by manipulating their bids in the case of an auction) to game the system and achieve a more favorable outcome. In this paper, we study such incentive-aware learning problem in a general setting and show that it is possible to approximately optimize the objective function under two assumptions: (i) each individual agent is a "small" (part of the market); and (ii) there is a cost associated with manipulation. For our illustrative application, this nicely translates to a mechanism for setting approximately optimal reserve prices in auctions where no individual agent has significant market share. For this application, we also show that the second assumption (that manipulations are costly) is not necessary since we can "perturb" any auction to make it costly for the agents to manipulate.
- Kareem Amin, Afshin Rostamizadeh, and Umar Syed. 2013. Learning prices for repeated auctions with strategic buyers Advances in Neural Information Processing Systems. 1169--1177. Google ScholarDigital Library
- Kareem Amin, Afshin Rostamizadeh, and Umar Syed. 2014. Repeated contextual auctions with strategic buyers Advances in Neural Information Processing Systems. 622--630. Google ScholarDigital Library
- M-F Balcan, Avrim Blum, Jason D Hartline, and Yishay Mansour. 2005. Mechanism design via machine learning. In Foundations of Computer Science, 2005. FOCS 2005. 46th Annual IEEE Symposium on. IEEE, 605--614. Google ScholarDigital Library
- Maria-Florina Balcan, Tuomas Sandholm, and Ellen Vitercik. 2017. Sample Complexity of Multi-Item Profit Maximization. arXiv preprint arXiv:1705.00243 (2017).Google Scholar
- Eric Budish. 2011. The combinatorial assignment problem: Approximate competitive equilibrium from equal incomes. Journal of Political Economy Vol. 119, 6 (2011), 1061--1103.Google ScholarCross Ref
- Nicolo Cesa-Bianchi, Claudio Gentile, and Yishay Mansour. 2015. Regret minimization for reserve prices in second-price auctions. IEEE Transactions on Information Theory Vol. 61, 1 (2015), 549--564.Google ScholarCross Ref
- Shuchi Chawla, Jason Hartline, and Denis Nekipelov. 2014. Mechanism design for data science. In Proceedings of the fifteenth ACM conference on Economics and computation. ACM, 711--712. Google ScholarDigital Library
- Richard Cole and Tim Roughgarden. 2014. The sample complexity of revenue maximization. In Proceedings of the 46th Annual ACM Symposium on Theory of Computing. ACM, 243--252. Google ScholarDigital Library
- Ofer Dekel, Felix Fischer, and Ariel D Procaccia. 2010. Incentive compatible regression learning. J. Comput. System Sci. Vol. 76, 8 (2010), 759--777. Google ScholarDigital Library
- Nikhil R Devanur, Zhiyi Huang, and Christos-Alexandros Psomas. 2015. The sample complexity of auctions with side information. arXiv preprint arXiv:1511.02296 (2015). Google ScholarDigital Library
- Piotr Faliszewski, Edith Hemaspaandra, and Lane A Hemaspaandra. 2010. Using complexity to protect elections. Commun. ACM Vol. 53, 11 (2010), 74--82. Google ScholarDigital Library
- Piotr Faliszewski and Ariel D Procaccia. 2010. AI's war on manipulation: Are we winning? AI Magazine Vol. 31, 4 (2010), 53--64.Google ScholarCross Ref
- Amos Fiat, Anna Karlin, Elias Koutsoupias, and Angelina Vidali. 2013. Approaching utopia: strong truthfulness and externality-resistant mechanisms Proceedings of the 4th conference on Innovations in Theoretical Computer Science. ACM, 221--230. Google ScholarDigital Library
- Hu Fu, Nicole Immorlica, Brendan Lucier, and Philipp Strack. 2015. Randomization beats second price as a prior-independent auction Proceedings of the Sixteenth ACM Conference on Economics and Computation. ACM, 323--323. Google ScholarDigital Library
- Yannai A Gonczarowski and Noam Nisan. 2016. Efficient Empirical Revenue Maximization in Single-Parameter Auction Environments. arXiv preprint arXiv:1610.09976 (2016).Google Scholar
- Moritz Hardt, Nimrod Megiddo, Christos Papadimitriou, and Mary Wootters. 2016. Strategic classification. In Proceedings of the 2016 ACM Conference on Innovations in Theoretical Computer Science. ACM, 111--122. Google ScholarDigital Library
- Zhiyi Huang, Yishay Mansour, and Tim Roughgarden. 2015. Making the most of your samples. In Proceedings of the Sixteenth ACM Conference on Economics and Computation. ACM, 45--60. Google ScholarDigital Library
- Yash Kanoria and Hamid Nazerzadeh. 2014. Dynamic reserve prices for repeated auctions: Learning from bids. (2014).Google Scholar
- Michael Kearns, Mallesh Pai, Aaron Roth, and Jonathan Ullman. 2014. Mechanism design in large games: Incentives and privacy Proceedings of the 5th conference on Innovations in theoretical computer science. ACM, 403--410. Google ScholarDigital Library
- Fuhito Kojima and Parag A Pathak. 2009. Incentives and stability in large two-sided matching markets. The American Economic Review Vol. 99, 3 (2009), 608--627.Google ScholarCross Ref
- Pinyan Lu, Shang-Hua Teng, and Changyuan Yu. 2006. Truthful auctions with optimal profit. In International Workshop on Internet and Network Economics. Springer, 27--36. Google ScholarDigital Library
- Frank McSherry and Kunal Talwar. 2007. Mechanism design via differential privacy. In Foundations of Computer Science, 2007. FOCS'07. 48th Annual IEEE Symposium on. IEEE, 94--103. Google ScholarDigital Library
- Reshef Meir, Ariel D Procaccia, and Jeffrey S Rosenschein. 2012. Algorithms for strategyproof classification. Artificial Intelligence Vol. 186 (2012), 123--156.Google ScholarDigital Library
- Jamie H Morgenstern and Tim Roughgarden. 2015. On the pseudo-dimension of nearly optimal auctions Advances in Neural Information Processing Systems. 136--144. Google ScholarDigital Library
- Noam Nisan, Tim Roughgarden, Eva Tardos, and Vijay V Vazirani. 2007. Algorithmic game theory. Vol. Vol. 1. Cambridge University Press Cambridge. Google ScholarDigital Library
- Kobbi Nissim, Rann Smorodinsky, and Moshe Tennenholtz. 2012. Approximately optimal mechanism design via differential privacy Proceedings of the 3rd innovations in theoretical computer science conference. ACM, 203--213. Google ScholarDigital Library
- Michael Ostrovsky and Michael Schwarz. 2011. Reserve prices in internet advertising auctions: A field experiment Proceedings of the 12th ACM conference on Electronic commerce. ACM, 59--60. Google ScholarDigital Library
- Pingzhong Tang and Yulong Zeng. 2016. How to manipulate truthful prior-dependent mechanisms? CoRR Vol. abs/1606.02409 (2016). deftempurl%http://arxiv.org/abs/1606.02409 tempurlGoogle Scholar
Index Terms
- Incentive-Aware Learning for Large Markets
Recommendations
Testing Incentive Compatibility in Display Ad Auctions
WWW '18: Proceedings of the 2018 World Wide Web ConferenceConsider a buyer participating in a repeated auction, such as those prevalent in display advertising. How would she test whether the auction is incentive compatible? To bid effectively, she is interested in whether the auction is single-shot incentive ...
Multiplicative Pacing Equilibria in Auction Markets
Budgets play a significant role in ad markets that implement sequential auctions such as those hosted by internet companies. In “Multiplicative Pacing Equilibria in Auction Markets,” the authors look at pacing in an ad marketplace using the lens of game ...
Budgets play a significant role in real-world sequential auction markets such as those implemented by internet companies. To maximize the value provided to auction participants, spending is smoothed across auctions so budgets are used for the best ...
Ex-post IR Dynamic Auctions with Cost-per-action Payments
AAMAS '18: Proceedings of the 17th International Conference on Autonomous Agents and MultiAgent SystemsConsider a repeated auction between one seller and many buyers, where each buyer only has an estimation of her value in each period until she actually receives the item in that period. The seller is allowed to conduct a dynamic auction to sell the items ...
Comments