Learning and Collusion in Multi-unit Auctions

Simina Branzei, Mahsa Derakhshan, Negin Golrezaei, Yanjun Han

Keywords: multi-unit auctions, repeated auctions, online learning, collusion, games and learning, lower bounds, multiplicative weight updates, bandit learning

TL;DR: We consider repeated multi-unit auctions with uniform pricing and study offline/online learning and equilibrium quality in two main auction variants, finding that one variant is susceptible to collusion among bidders while the other is not.

Abstract: In a carbon auction, licenses for CO2 emissions are allocated among multiple interested players. Inspired by this setting, we consider repeated multi-unit auctions with uniform pricing, which are widely used in practice. Our contribution is to analyze these auctions in both the offline and online settings, by designing efficient bidding algorithms with low regret and giving regret lower bounds. We also analyze the quality of the equilibria in two main variants of the auction, finding that one variant is susceptible to collusion among the bidders while the other is not.

Submission Number: 2857