Diffusion models enable multiple winners in banner ad auctions
Diffusion-Based Auctions for Multi-Winner Banner Ads
Computer Science and Game TheoryArtificial IntelligenceMachine Learning
Summary
Banner ads are usually sold to just one advertiser who shows their single ad. This paper shows how an auction method using diffusion models can let several advertisers share one banner and influence the final image based on their bids. Advertisers describe what they want in words, and the system uses these descriptions and bids to create an ad that balances their preferences. The method was tested and found to be more efficient, fair, and keeps the ad image quality good.
What this means in practice
- •For digital advertising teams: Enable multiple advertisers to share a single banner ad with dynamic image blending driven by their bids for richer ad experiences.$Commercial implications: Supports selling multi-ad banner slots that combine advertiser preferences into a single image, creating new ad space products.
- •For online marketplace operators: Adopt auction mechanisms that increase auction efficiency and fairness when allocating banner ad space among multiple bidders simultaneously.
Authors
Lilian Sun, Henry Huang, Fucheng Warren Zhu, Giannis Daras, Constantinos Daskalakis
Abstract
We propose a new framework for banner ad auctions, leveraging diffusion models to allow for multiple bidders to co-win a single banner ad opportunity, and in particular influence the ad shown on the banner depending on the strength of their bids, enabling a far richer outcome space compared to standard banner ad auctions, where there is a single winner who shows their ad. In our framework, the outcomes are images (what ad to show) and each bidder expresses their preferences over outcomes by declaring their value per unit of alignment between the image chosen by the mechanism and a textual description of their desired image. Our proposed mechanism combines the advertisers' bids and textual descriptions into a score that drives the reverse process of a diffusion model. This is used to sample a set of candidate images from which the welfare maximizing one is eventually chosen. Extensive experimentation evaluates our proposed mechanism around the following axes: (1) efficiency improvements of our allocation rule relative to the single-winner allocation rule; (2) monotonicity of our allocation rule; and (3) approximate incentive compatibility when combining our allocation rule with Myerson lemma-based or Clarke rule-based payments. Improvements along all these axes are shown with the number of Monte Carlo samples. Meanwhile, these benefits are attained while preserving the image quality.