Optimal Auction Design in the Joint Advertising
This addresses the problem of inefficient revenue generation in online advertising platforms by improving auction design for joint advertising, representing an incremental advance with a novel method for a known bottleneck.
The paper tackles the problem of designing optimal auction mechanisms for joint advertising, where bundles of two advertisers are assigned to ad slots, and finds that existing mechanisms are suboptimal due to overlooking bundle structures. It proposes BundleNet, a bundle-based neural network approach for multi-slot settings, which approximates theoretical optimality in single-slot cases and achieves state-of-the-art performance in multi-slot settings, significantly increasing platform revenue while ensuring approximate incentive compatibility and individual rationality.
Online advertising is a vital revenue source for major internet platforms. Recently, joint advertising, which assigns a bundle of two advertisers in an ad slot instead of allocating a single advertiser, has emerged as an effective method for enhancing allocation efficiency and revenue. However, existing mechanisms for joint advertising fail to realize the optimality, as they tend to focus on individual advertisers and overlook bundle structures. This paper identifies an optimal mechanism for joint advertising in a single-slot setting. For multi-slot joint advertising, we propose \textbf{BundleNet}, a novel bundle-based neural network approach specifically designed for joint advertising. Our extensive experiments demonstrate that the mechanisms generated by \textbf{BundleNet} approximate the theoretical analysis results in the single-slot setting and achieve state-of-the-art performance in the multi-slot setting. This significantly increases platform revenue while ensuring approximate dominant strategy incentive compatibility and individual rationality.