Learning & Games Reading Group: Learning in the Presence of Strategic Behavior

Title: Sample complexity for non-truthful mechanisms

Authors: Jason Hartline and Sam Taggart

Link: https://arxiv.org/abs/1608.01875

Abstract: This paper considers the design of non-truthful mechanisms from samples. We identify a parameterized family of mechanisms with strategically simple winner-pays-bid, all-pay, and truthful payment formats. In general (not necessarily downward-closed) single-parameter feasibility environments we prove that the family has low representation and generalization error. Specifically, polynomially many bid samples suffice to identify and run a mechanism that is ε-close in Bayes-Nash equilibrium revenue or welfare to that of the optimal truthful mechanism.