Online Platt Scaling with Calibeating

Chirag Gupta, Aaditya Ramdas
Proceedings of the 40th International Conference on Machine Learning, PMLR 202:12182-12204, 2023.

Abstract

We present an online post-hoc calibration method, called Online Platt Scaling (OPS), which combines the Platt scaling technique with online logistic regression. We demonstrate that OPS smoothly adapts between i.i.d. and non-i.i.d. settings with distribution drift. Further, in scenarios where the best Platt scaling model is itself miscalibrated, we enhance OPS by incorporating a recently developed technique called calibeating to make it more robust. Theoretically, our resulting OPS+calibeating method is guaranteed to be calibrated for adversarial outcome sequences. Empirically, it is effective on a range of synthetic and real-world datasets, with and without distribution drifts, achieving superior performance without hyperparameter tuning. Finally, we extend all OPS ideas to the beta scaling method.

Cite this Paper

BibTeX
@InProceedings{pmlr-v202-gupta23c, title = {Online Platt Scaling with Calibeating}, author = {Gupta, Chirag and Ramdas, Aaditya}, booktitle = {Proceedings of the 40th International Conference on Machine Learning}, pages = {12182--12204}, year = {2023}, editor = {Krause, Andreas and Brunskill, Emma and Cho, Kyunghyun and Engelhardt, Barbara and Sabato, Sivan and Scarlett, Jonathan}, volume = {202}, series = {Proceedings of Machine Learning Research}, month = {23--29 Jul}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v202/gupta23c/gupta23c.pdf}, url = {https://proceedings.mlr.press/v202/gupta23c.html}, abstract = {We present an online post-hoc calibration method, called Online Platt Scaling (OPS), which combines the Platt scaling technique with online logistic regression. We demonstrate that OPS smoothly adapts between i.i.d. and non-i.i.d. settings with distribution drift. Further, in scenarios where the best Platt scaling model is itself miscalibrated, we enhance OPS by incorporating a recently developed technique called calibeating to make it more robust. Theoretically, our resulting OPS+calibeating method is guaranteed to be calibrated for adversarial outcome sequences. Empirically, it is effective on a range of synthetic and real-world datasets, with and without distribution drifts, achieving superior performance without hyperparameter tuning. Finally, we extend all OPS ideas to the beta scaling method.} }
Endnote
%0 Conference Paper %T Online Platt Scaling with Calibeating %A Chirag Gupta %A Aaditya Ramdas %B Proceedings of the 40th International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2023 %E Andreas Krause %E Emma Brunskill %E Kyunghyun Cho %E Barbara Engelhardt %E Sivan Sabato %E Jonathan Scarlett %F pmlr-v202-gupta23c %I PMLR %P 12182--12204 %U https://proceedings.mlr.press/v202/gupta23c.html %V 202 %X We present an online post-hoc calibration method, called Online Platt Scaling (OPS), which combines the Platt scaling technique with online logistic regression. We demonstrate that OPS smoothly adapts between i.i.d. and non-i.i.d. settings with distribution drift. Further, in scenarios where the best Platt scaling model is itself miscalibrated, we enhance OPS by incorporating a recently developed technique called calibeating to make it more robust. Theoretically, our resulting OPS+calibeating method is guaranteed to be calibrated for adversarial outcome sequences. Empirically, it is effective on a range of synthetic and real-world datasets, with and without distribution drifts, achieving superior performance without hyperparameter tuning. Finally, we extend all OPS ideas to the beta scaling method.
APA
Gupta, C. & Ramdas, A.. (2023). Online Platt Scaling with Calibeating. Proceedings of the 40th International Conference on Machine Learning, in Proceedings of Machine Learning Research 202:12182-12204 Available from https://proceedings.mlr.press/v202/gupta23c.html.

Related Material