Cost-sensitive Learning for Bidding in Online Advertising Auctions

One of the most challenging problems in computational advertising is the prediction of ad click and conversion rates for bidding in online advertising auctions. State-ofthe-art prediction methods include using the maximum entropy framework (also known as logistic regression) and log linear models. However, one unaddressed problem in the previous approaches is the existence of highly non-uniform misprediction costs. In this paper, we present our approach for making cost-sensitive predictions for bidding in online advertising auctions. We show that one can get significant lifts in offline and online performance by using a simple modification of the logistic loss function.