Data-Driven Mechanism Design: Jointly Eliciting Preferences and Information

TL;DR

This paper introduces data-driven mechanisms that improve implementation of efficient allocations in environments with private preferences and information, using estimators of payoff-relevant states to extend classic auction frameworks.

Contribution

It proposes novel data-driven mechanisms that condition transfers on post-allocation information, achieving exact or approximate implementation in complex multi-dimensional settings.

Findings

Mechanisms achieve exact implementation with fully revealed states or affine utilities.

Approximate implementation converges to exact as estimators improve.

Applications demonstrated in digital advertising, AI shopping, and procurement auctions.

Abstract

We study mechanism design in environments where agents have private preferences and private information about a common payoff-relevant state. In such settings with multi-dimensional types, standard mechanisms fail to implement efficient allocations. We address this limitation by proposing data-driven mechanisms that condition transfers on additional post-allocation information, modeled as an estimator of the payoff-relevant state. Our mechanisms extend the classic Vickrey-Clarke-Groves framework. We show they achieve exact implementation in posterior equilibrium when the state is fully revealed or utilities are affine in an unbiased estimator. With a consistent estimator, they achieve approximate implementation that converges to exact implementation as the estimator converges, and we provide bounds on the convergence rate. We demonstrate applications to digital advertising auctions and AI shopping assistants, where user engagement naturally reveals relevant information, and to procurement auctions with consumer spot markets, where additional information arises from a pricing game played by the same agents.