Abstract: Machine learning models are susceptible to a variety of attacks that can erode trust, including attacks against the privacy of training data, and adversarial examples that jeopardize model accuracy. Differential privacy and certified robustness are effective frameworks for combating these two threats respectively, as they each provide future-proof guarantees. However, we show that standard differentially private model training is insufficient for providing strong certified robustness guarantees. Indeed, combining differential privacy and certified robustness in a single system is non-trivial, leading previous works to introduce complex training schemes that lack flexibility. In this work, we present DP-CERT, a simple and effective method that achieves both privacy and robustness guarantees simultaneously by integrating randomized smoothing into standard differentially private model training. Compared to the leading prior work, DP-CERT gives up to a 2.5x increase in certified accuracy for the same differential privacy guarantee on CIFAR10. Through in-depth per-sample metric analysis, we find that larger certifiable radii correlate with smaller local Lipschitz constants, and show that DP-CERT effectively reduces Lipschitz constants compared to other differentially private training methods. Code is available at github.com/layer6ai-labs/dp-cert.
Submission Length: Regular submission (no more than 12 pages of main content)
Changes Since Last Submission: Camera ready revision
---
Added tables with more hyperparameter ablations and latency results.
Added detailed explanation of privacy guarantee and the relationship between Differential Privacy and Certified Robustness.
Emphasized that the experiments are conducted under the same epsilon and delta parameters.
Code: https://github.com/layer6ai-labs/dp-cert
Assigned Action Editor: ~Yiming_Ying1
Submission Number: 2287
Loading