TEASE: A Leak-Resilient Strong PUFs Construction via Statistically Deficient Data Release
Summary
PUFs (Physical Unclonable Functions, hardware devices that generate unique digital fingerprints) are vulnerable to machine learning attacks that can predict their responses if attackers obtain leaked challenge-response pairs (CRPs, inputs and outputs used to test PUFs). This paper presents the TEASE algorithm, which selects CRPs more strategically instead of randomly so that even if attackers leak them, the algorithm forces machine learning models to achieve only about 50% accuracy, comparable to random guessing, while resisting multiple rounds of leaks and powerful attacks.
Solution / Mitigation
The TEASE algorithm addresses the vulnerability by using bilevel programming (an optimization technique that optimizes one problem while considering another problem inside it) to carefully select which challenge-response pairs to allow in circulation. According to the source, TEASE 'forces ML attacks to be content with a prediction accuracy of around 50% on non-leaked CRPs, comparable to random guessing' and 'can gracefully tolerate multiple rounds of CRP leaks' while maintaining 'lower hardware overhead than existing countermeasures.'
Classification
Original source: http://ieeexplore.ieee.org/document/11570910
First tracked: July 24, 2026 at 08:03 PM
Classified by LLM (prompt v3) · confidence: 85%