Accuracy, Interpretability, and Differential Privacy via Explainable Boosting
Abstract
We show that adding differential privacy to Explainable Boosting Machines (EBMs), a recent method for training interpretable ML models, yields state-of-the-art accuracy while protecting privacy. Our experiments on multiple classification and regression datasets show that DP-EBM models suffer surprisingly little accuracy loss even with strong differential privacy guarantees. In addition to high accuracy, two other benefits of applying DP to EBMs are: a) trained models provide exact global and local interpretability, which is often important in settings where differential privacy is needed; and b) the models can be edited after training without loss of privacy to correct errors which DP noise may have introduced.
Cite
@article{arxiv.2106.09680,
title = {Accuracy, Interpretability, and Differential Privacy via Explainable Boosting},
author = {Harsha Nori and Rich Caruana and Zhiqi Bu and Judy Hanwen Shen and Janardhan Kulkarni},
journal= {arXiv preprint arXiv:2106.09680},
year = {2021}
}
Comments
To be published in ICML 2021. 12 pages, 6 figures