Posterior Probability Matters: Doubly-Adaptive Calibration for Neural Predictions in Online Advertising
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2022
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866914810503888896 |
|---|---|
| author | Wei, Penghui Zhang, Weimin Hou, Ruijie Liu, Jinquan Liu, Shaoguo Wang, Liang Zheng, Bo |
| author_facet | Wei, Penghui Zhang, Weimin Hou, Ruijie Liu, Jinquan Liu, Shaoguo Wang, Liang Zheng, Bo |
| contents | 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. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2205_07295 |
| institution | arXiv |
| publishDate | 2022 |
| record_format | arxiv |
| spellingShingle | Posterior Probability Matters: Doubly-Adaptive Calibration for Neural Predictions in Online Advertising Wei, Penghui Zhang, Weimin Hou, Ruijie Liu, Jinquan Liu, Shaoguo Wang, Liang Zheng, Bo Machine Learning Information Retrieval 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. |
| title | Posterior Probability Matters: Doubly-Adaptive Calibration for Neural Predictions in Online Advertising |
| topic | Machine Learning Information Retrieval |
| url | https://arxiv.org/abs/2205.07295 |