Unbiased Lift-based Bidding System
This addresses the profitability issue for advertisers in online display ad auctions by enabling more accurate bidding strategies, though it appears incremental as it builds on existing lift-based approaches.
The paper tackles the problem of biased log data in online ad auctions by developing an unbiased lift-based bidding system that predicts the lift-effect of showing ads to maximize advertiser profit, demonstrating its superiority through real-world A/B testing.
Conventional bidding strategies for online display ad auction heavily relies on observed performance indicators such as clicks or conversions. A bidding strategy naively pursuing these easily observable metrics, however, fails to optimize the profitability of the advertisers. Rather, the bidding strategy that leads to the maximum revenue is a strategy pursuing the performance lift of showing ads to a specific user. Therefore, it is essential to predict the lift-effect of showing ads to each user on their target variables from observed log data. However, there is a difficulty in predicting the lift-effect, as the training data gathered by a past bidding strategy may have a strong bias towards the winning impressions. In this study, we develop Unbiased Lift-based Bidding System, which maximizes the advertisers' profit by accurately predicting the lift-effect from biased log data. Our system is the first to enable high-performing lift-based bidding strategy by theoretically alleviating the inherent bias in the log. Real-world, large-scale A/B testing successfully demonstrates the superiority and practicability of the proposed system.