Take a look at the new version of this page: [ beta version ]. Tell us what you think.

ABSTRACT

Much of the research in auction theory assumes that the auctioneer knows the distribution of participants' valuations with complete certainty. However, this is unrealistic. Thus, we analyse cases in which the auctioneer is uncertain about the valuation distributions; specifically, we consider a repeated auction setting in which the auctioneer can learn these distributions. Using take-it-or-leave-it auctions (Sandholm and Gilpin, 2006) as an exemplar auction format, we consider two auction design criteria. Firstly, an auctioneer could maximise expected revenue each time the auction is held. Secondly, an auctioneer could maximise the information gained in earlier auctions (as measured by the Kullback-Liebler divergence between its posterior and prior) to develop good estimates of the unknowns, which are later exploited to improve the revenue earned in the long-run. Simulation results comparing the two criteria indicate that setting offers to maximise revenue does not significantly detract from learning performance, but optimising offers for information gain substantially reduces expected revenue while not producing significantly better parameter estimates.

REFERENCES

• 1: J. M. Bernardo. Expected information as expected utility. The Annals of Statistics, 7:686--690, 1979.
• 2: J. Bredin and D. C. Parkes. Models for truthful online double auctions. In Proceeding of the 21th Annual Conference on Uncertainty in Artificial Intelligence (UAI-05), pages 50--59, Arlington, Virginia, 2005. AUAI Press.
• 3: Vincent Conitzer , Nikesh Garera, Learning algorithms for online principal-agent problems (and selling goods online), Proceedings of the 23rd international conference on Machine learning, p.209-216, June 25-29, 2006, Pittsburgh, Pennsylvania [doi>10.1145/1143844.1143871]
• 4: David J. C. MacKay, Information Theory, Inference & Learning Algorithms, Cambridge University Press, New York, NY, 2002
• 5: R. Myerson. Optimal auction design. Mathematics of Operations Research, 6(1):58--73, 1981.
• 6: David Pardoe , Peter Stone , Maytal Saar-Tsechansky , Kerem Tomak, Adaptive mechanism design: a metalearning approach, Proceedings of the 8th international conference on Electronic commerce: The new e-commerce: innovations for conquering current barriers, obstacles and limitations to conducting successful business on the internet, August 13-16, 2006, Fredericton, New Brunswick, Canada [doi>10.1145/1151454.1151480]
• 7: A. Rogers, E. David, J. Schiff, S. Kraus, and N. R. Jennings. Learning environmental parameters for the design of optimal English auctions with discrete bid levels. In H. L. Poutré, N. Sadeh, and J. Sverker, editors, Agent-mediated Electronic Commerce, Designing Trading Agents and Mechanisms: AAMAS 2005 Workshop, AMEC, pages 1--15, Utrecht, Netherlands, July 25 2005. Springer.
• 8: Tuomas Sandholm , Andrew Gilpin, Sequences of take-it-or-leave-it offers: near-optimal auctions without full valuation revelation, Proceedings of the fifth international joint conference on Autonomous agents and multiagent systems, May 08-12, 2006, Hakodate, Japan [doi>10.1145/1160633.1160839]

INDEX TERMS

Primary Classification:
  J. Computer Applications
  J.4 SOCIAL AND BEHAVIORAL SCIENCES

General Terms:
Design, Economics

Keywords:
Bayesian experimental design, optimal auction design