Reserve Price Optimization for First Price Auctions
This addresses a practical problem for publishers in the display advertising industry who need to adapt to the transition from second- to first-price auctions, though it represents an incremental improvement over existing optimization techniques.
The paper tackles the problem of optimizing reserve prices in first-price auctions for display advertising, proposing a gradient-based algorithm that reduces variance in gradient estimates to improve convergence rates. The algorithm achieves significant revenue improvements, with experiments showing up to 15% revenue gains on real Google ad exchange data compared to baseline methods.
The display advertising industry has recently transitioned from second- to first-price auctions as its primary mechanism for ad allocation and pricing. In light of this, publishers need to re-evaluate and optimize their auction parameters, notably reserve prices. In this paper, we propose a gradient-based algorithm to adaptively update and optimize reserve prices based on estimates of bidders' responsiveness to experimental shocks in reserves. Our key innovation is to draw on the inherent structure of the revenue objective in order to reduce the variance of gradient estimates and improve convergence rates in both theory and practice. We show that revenue in a first-price auction can be usefully decomposed into a \emph{demand} component and a \emph{bidding} component, and introduce techniques to reduce the variance of each component. We characterize the bias-variance trade-offs of these techniques and validate the performance of our proposed algorithm through experiments on synthetic data and real display ad auctions data from Google ad exchange.