Cost-sensitive Learning for Utility Optimization in Online Advertising Auctions
This work addresses a specific problem for advertisers in computational advertising by improving utility optimization, but it is incremental as it builds on existing log loss methods.
The paper tackles the problem of non-uniform misprediction costs in online advertising auctions by analyzing the relationship between optimizing advertiser profit (Utility metric) and log loss, proposing a cost weighting scheme for log loss. It shows significant gains in offline and online performance, though specific numbers are not provided.
One of the most challenging problems in computational advertising is the prediction of click-through and conversion rates for bidding in online advertising auctions. An unaddressed problem in previous approaches is the existence of highly non-uniform misprediction costs. While for model evaluation these costs have been taken into account through recently proposed business-aware offline metrics -- such as the Utility metric which measures the impact on advertiser profit -- this is not the case when training the models themselves. In this paper, to bridge the gap, we formally analyze the relationship between optimizing the Utility metric and the log loss, which is considered as one of the state-of-the-art approaches in conversion modeling. Our analysis motivates the idea of weighting the log loss with the business value of the predicted outcome. We present and analyze a new cost weighting scheme and show that significant gains in offline and online performance can be achieved.