Machine Learning-powered Iterative Combinatorial Auctions

We present a machine learning-powered iterative combinatorial auction (CA). The main goal of integrating machine learning (ML) into the auction is to improve preference elicitation, which is a major challenge in large CAs. In contrast to prior work, our auction design uses value queries instead of prices to drive the auction. The ML algorithm is used to help the auction decide which value queries to ask in every iteration. While using ML inside a CA introduces new challenges, we demonstrate how we obtain a design that is individually rational, has good incentives, and is computationally practical. We benchmark our new auction against the well-known combinatorial clock auction (CCA). Our results indicate that, especially in large domains, our ML-powered auction can achieve higher allocative efficiency than the CCA, even with only a small number of value queries.

[1]  Ailsa H. Land,et al.  An Automatic Method of Solving Discrete Programming Problems , 1960 .

[2]  Tim Roughgarden,et al.  On the Pseudo-Dimension of Nearly Optimal Auctions , 2015, NIPS.

[3]  Paul Dütting,et al.  Payment Rules through Discriminant-Based Classifiers , 2012, ACM Trans. Economics and Comput..

[4]  Lawrence M. Ausubel,et al.  A Practical Guide to the Combinatorial Clock Auction , 2017 .

[5]  Vitali Gretschko,et al.  Strategic Complexities in the Combinatorial Clock Auction , 2016, SSRN Electronic Journal.

[6]  Lawrence M. Ausubel,et al.  The Clock-Proxy Auction: A Practical Combinatorial Auction Design , 2004 .

[7]  Noam Nisan,et al.  Computationally feasible VCG mechanisms , 2000, EC '00.

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

[9]  Tim Roughgarden,et al.  Learning Simple Auctions , 2016, COLT.

[10]  Sven Seuken,et al.  Fast Iterative Combinatorial Auctions via Bayesian Learning , 2018, AAAI.

[11]  Don R. Hush,et al.  Learning from dependent observations , 2007, J. Multivar. Anal..

[12]  Maria-Florina Balcan,et al.  A General Theory of Sample Complexity for Multi-Item Profit Maximization , 2017, EC.

[13]  Avrim Blum,et al.  Preference Elicitation and Query Learning , 2004, J. Mach. Learn. Res..

[14]  Tuomas Sandholm,et al.  Preference elicitation in combinatorial auctions , 2001, AAMAS '02.

[15]  Maria-Florina Balcan,et al.  Sample Complexity of Multi-Item Profit Maximization , 2017, ArXiv.

[16]  Noam Nisan,et al.  The communication requirements of efficient allocations and supporting prices , 2006, J. Econ. Theory.

[17]  Noam Nisany,et al.  The Communication Requirements of E¢cient Allocations and Supporting Lindahl Prices¤ , 2003 .

[18]  Maarten C. W. Janssen,et al.  On the Clock of the Combinatorial Clock Auction , 2016 .

[19]  Vincent Conitzer,et al.  Self-interested automated mechanism design and implications for optimal combinatorial auctions , 2004, EC '04.

[20]  Vincent Conitzer,et al.  Complexity of Mechanism Design , 2002, UAI.

[21]  David C. Parkes,et al.  Applying learning algorithms to preference elicitation , 2004, EC '04.

[22]  David C. Parkes,et al.  Deep Learning for Multi-Facility Location Mechanism Design , 2018, IJCAI.

[23]  David C. Parkes,et al.  ICE: An Expressive Iterative Combinatorial Exchange , 2008, J. Artif. Intell. Res..

[24]  Robert W. Day,et al.  Quadratic Core-Selecting Payment Rules for Combinatorial Auctions , 2012, Oper. Res..

[25]  A. Land,et al.  An Automatic Method for Solving Discrete Programming Problems , 1960, 50 Years of Integer Programming.

[26]  Martin Bichler,et al.  Compact Bid Languages and Core Pricing in Large Multi-item Auctions , 2015, Manag. Sci..

[27]  Sébastien Lahaie,et al.  A Kernel-Based Iterative Combinatorial Auction , 2011, AAAI.

[28]  Carl E. Rasmussen,et al.  Gaussian processes for machine learning , 2005, Adaptive computation and machine learning.

[29]  Sébastien Lahaie,et al.  A Bayesian Clearing Mechanism for Combinatorial Auctions , 2017, AAAI.

[30]  David C. Parkes,et al.  Iterative Combinatorial Auctions , 2006 .

[31]  Noam Nissan,et al.  Bidding Languages , 2004 .

[32]  Adam N. Letchford,et al.  On Nonconvex Quadratic Programming with Box Constraints , 2009, SIAM J. Optim..

[33]  Benjamin Lubin,et al.  Adaptive-Price Combinatorial Auctions , 2018, EC.

[34]  Jonathan D. Levin,et al.  Properties of the Combinatorial Clock Auction , 2016 .

[35]  Peter Cramton,et al.  Auction Design for Wind Rights , 2012 .

[36]  Ignacio E. Grossmann,et al.  On the solution of nonconvex cardinality Boolean quadratic programming problems: a computational study , 2017, Comput. Optim. Appl..

[37]  Jacob K. Goeree,et al.  Hierarchical package bidding: A paper & pencil combinatorial auction , 2010, Games Econ. Behav..

[38]  Richard Cole,et al.  The sample complexity of revenue maximization , 2014, STOC.

[39]  M. Bichler,et al.  Do core-selecting Combinatorial Clock Auctions always lead to high efficiency? An experimental analysis of spectrum auction designs , 2013 .

[40]  Sven de Vries,et al.  On ascending Vickrey auctions for heterogeneous objects , 2007, J. Econ. Theory.

[41]  David C. Parkes,et al.  An Iterative Generalized Vickrey Auction: Strategy-Proofness without Complete Revelation , 2001 .

[42]  Sushil Bikhchandani,et al.  An Efficient Dynamic Auction for Heterogeneous Commodities , 2006 .

[43]  J. Goeree,et al.  Handbook of Spectrum Auction Design , 2017 .

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

[45]  Claudio Gentile,et al.  Ieee Transactions on Information Theory 1 Regret Minimization for Reserve Prices in Second-price Auctions , 2022 .

[46]  Bernhard Schölkopf,et al.  A tutorial on support vector regression , 2004, Stat. Comput..

[47]  Vasilis Syrgkanis.  A Sample Complexity Measure with Applications to Learning Optimal Auctions , 2017, NIPS.

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

[49]  David C. Parkes,et al.  Automated Mechanism Design without Money via Machine Learning , 2016, IJCAI.

[50]  Mehryar Mohri,et al.  Learning Theory and Algorithms for revenue optimization in second price auctions with reserve , 2013, ICML.

[51]  Sven Seuken,et al.  Combinatorial Auctions via Machine Learning-based Preference Elicitation , 2018, IJCAI.

[52]  Sven Seuken,et al.  Deep Learning-powered Iterative Combinatorial Auctions , 2019, AAAI.

[53]  A. Mas-Colell,et al.  Microeconomic Theory , 1995 .

[54]  Vincent Conitzer,et al.  Combinatorial Auctions with k-wise Dependent Valuations , 2005, AAAI.

[55]  Sven Seuken,et al.  Fourier Analysis-based Iterative Combinatorial Auctions , 2020, ArXiv.

[56]  Paul Dütting,et al.  Optimal auctions through deep learning , 2017, ICML.

[57]  David C. Parkes,et al.  Ascending Price Vickrey Auctions for General Valuations , 2005, J. Econ. Theory.

[58]  Maria-Florina Balcan,et al.  Sample Complexity of Automated Mechanism Design , 2016, NIPS.

[59]  Martin Bichler,et al.  On the Impact of Cognitive Limits in Combinatorial Auctions: An Experimental Study in the Context of Spectrum Auction Design , 2010 .

[60]  J. Mercer.  Functions of Positive and Negative Type, and their Connection with the Theory of Integral Equations , 1909 .

[61]  A. N. Tikhonov,et al.  Solutions of ill-posed problems , 1977 .

[62]  T. Sandholm.  Very-Large-Scale Generalized Combinatorial Multi-Attribute Auctions: Lessons from Conducting $60 Billion of Sourcing , 2013 .

[63]  Tim Roughgarden,et al.  Ironing in the Dark , 2015, EC.

[64]  Sven Seuken,et al.  SATS: A Universal Spectrum Auction Test Suite , 2017, AAMAS.