Posterior Probability Matters: Doubly-Adaptive Calibration for Neural Predictions in Online Advertising
This work addresses the need for accurate probabilistic predictions in ad ranking and bidding, offering a practical solution for online advertising systems.
The paper tackles the problem of improving calibration of neural predictions for user response probabilities in online advertising by proposing AdaCalib, a doubly-adaptive approach that learns isotonic functions with field-adaptive mechanisms, resulting in significant calibration improvements and successful online deployment.
Predicting user response probabilities is vital for ad ranking and bidding. We hope that predictive models can produce accurate probabilistic predictions that reflect true likelihoods. Calibration techniques aim to post-process model predictions to posterior probabilities. Field-level calibration -- which performs calibration w.r.t. to a specific field value -- is fine-grained and more practical. In this paper we propose a doubly-adaptive approach AdaCalib. It learns an isotonic function family to calibrate model predictions with the guidance of posterior statistics, and field-adaptive mechanisms are designed to ensure that the posterior is appropriate for the field value to be calibrated. Experiments verify that AdaCalib achieves significant improvement on calibration performance. It has been deployed online and beats previous approach.