Zitao Chen (University of British Columbia), Karthik Pattabiraman (University of British Columbia)

Machine learning (ML) models are vulnerable to membership inference attacks (MIAs), which determine whether a given input is used for training the target model. While there have been many efforts to mitigate MIAs, they often suffer from limited privacy protection, large accuracy drop, and/or requiring additional data that may be difficult to acquire.

This work proposes a defense technique, HAMP that can achieve both strong membership privacy and high accuracy, without requiring extra data. To mitigate MIAs in different forms, we observe that they can be unified as they all exploit the ML model’s overconfidence in predicting training samples through different proxies. This motivates our design to enforce less confident prediction by the model, hence forcing the model to behave similarly on the training and testing samples. HAMP consists of a novel training framework with high-entropy soft labels and an entropy-based regularizer to constrain the model’s prediction while still achieving high accuracy. To further reduce privacy risk, HAMP uniformly modifies all the prediction outputs to become low-confidence outputs while preserving the accuracy, which effectively obscures the differences between the prediction on members and non-members.

We conduct extensive evaluation on five benchmark datasets, and show that HAMP provides consistently high accuracy and strong membership privacy. Our comparison with seven state-of- the-art defenses shows that HAMP achieves a superior privacy- utility trade off than those techniques.

View More Papers

TrustSketch: Trustworthy Sketch-based Telemetry on Cloud Hosts

Zhuo Cheng (Carnegie Mellon University), Maria Apostolaki (Princeton University), Zaoxing Liu (University of Maryland), Vyas Sekar (Carnegie Mellon University)

Read More

On Requirements and Concepts for TT&C Link Key Management

Christoph Bader (Airbus Defence & Space GmbH)

Read More

50 Shades of Support: A Device-Centric Analysis of Android...

Abbas Acar (Florida International University), Güliz Seray Tuncay (Google), Esteban Luques (Florida International University), Harun Oz (Florida International University), Ahmet Aris (Florida International University), Selcuk Uluagac (Florida International University)

Read More

Sticky Fingers: Resilience of Satellite Fingerprinting against Jamming Attacks

Joshua Smailes (University of Oxford), Edd Salkield (University of Oxford), Sebastian Köhler (University of Oxford), Simon Birnbach (University of Oxford), Martin Strohmeier (Cyber-Defence Campus, armasuisse S+T), Ivan Martinovic (University of Oxford)

Read More