# Cost-sensitive Learning for Utility Optimization in Online Advertising Auctions

arXiv: https://arxiv.org/abs/1603.03713 · HTML: https://mlsanity.com/paper/1603.03713/

## Key facts

- Cited by 7 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 11 Mar 2016.

## Built on by (most cited first, 7 of 7)

- [User Response Prediction in Online Advertising](https://mlsanity.com/paper/2101.02342/) (2021)
- [Pairwise Ranking Losses of Click-Through Rates Prediction for Welfare Maximization in Ad Auctions](https://mlsanity.com/paper/2306.01799/) (2023)
- [Model Monotonicity in Autobidding Auctions: When Do Better Predictions Lead to Better Outcomes?](https://mlsanity.com/paper/2605.31036/) (2026)
- [An Analysis of Selection Bias Issue for Online Advertising](https://mlsanity.com/paper/2206.03853/) (2022)
- [Risk-Aware Bid Optimization for Online Display Advertisement](https://mlsanity.com/paper/2210.15837/) (2022)
- [Confounding is a Pervasive Problem in Real World Recommender Systems](https://mlsanity.com/paper/2508.10479/) (2025)
- [Conversion rate prediction in online advertising: modeling techniques, performance evaluation and future directions](https://mlsanity.com/paper/2512.01171/) (2025)

## Abstract (arXiv)

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.