Comparing Position Auctions Computationally

Authors

  • David Thompson University of British Columbia
  • Kevin Leyton-Brown University of British Columbia

DOI:

https://doi.org/10.1609/aaai.v24i1.7710

Keywords:

Auctions and Market-Based Systems, E-Commerce, Game Theory

Abstract

Modern techniques for representing games and computing their Nash equilibria are approaching the point where they can be used to analyze market games. We demonstrate this by showing how the equilibria of different position auction mechanisms can be tractably identified using these techniques. These results enable detailed and quantitative comparisons of the different auction mechanisms — in terms of both efficiency and revenue — under different preference models and equilibrium selection criteria.

Downloads

Published

2010-07-05

How to Cite

Thompson, D., & Leyton-Brown, K. (2010). Comparing Position Auctions Computationally. Proceedings of the AAAI Conference on Artificial Intelligence, 24(1), 1694-1697. https://doi.org/10.1609/aaai.v24i1.7710

Issue

Section

New Scientific and Technical Advances in Research