Joint Auction in the Online Advertising Market

What is it about?

Online advertising is a primary source of income for e-commerce platforms. In the current advertising pattern, the oriented targets are the online store owners who are willing to pay extra fees to enhance the position of their stores. On the other hand, brand suppliers are also desirable to advertise their products in stores to boost brand sales. However, the currently used advertising mode cannot satisfy the demand of both stores and brand suppliers simultaneously. To address this, we innovatively propose a joint advertising model termed ``Joint Auction'', allowing brand suppliers and stores to collaboratively bid for advertising slots, catering to both their needs. However, conventional advertising auction mechanisms are not suitable for this novel scenario. In this paper, we propose JRegNet, a neural network architecture for the optimal joint auction design, to generate mechanisms that can achieve the optimal revenue and guarantee (near-)dominant strategy incentive compatibility and individual rationality. Finally, multiple experiments are conducted on synthetic and real data to demonstrate that our proposed joint auction significantly improves platform’s revenue compared to the known baselines.

Featured Image

Why is it important?

The advertising system's most pivotal technology is its sales mechanism. The industry has seen significant maturity in the traditional advertising auction mechanism, with most studies emphasizing revenue enhancement through the introduction of innovative variations to existing frameworks. Our study uniquely presents a practical and revenue-boosting advertising sales model known as the ``joint auction''. Nonetheless, many standard and widely-used mechanisms may not be applicable to this novel joint advertising model. To identify the optimal mechanisms that are both dominant strategy incentive compatible (DSIC) and individually rational (IR), we introduce JRegNet, a neural network architecture to generate the optimal satisfactory mechanisms. Subsequently, we validate our proposed architecture through numerous experiments on both synthetic and real-world datasets. These experiments demonstrate the superior performance of the generated mechanisms compared to established baselines.