Watch a 13-minute conference presentation from USENIX Security '24 exploring FAMOS, an innovative authentication framework for mobile payment applications. Learn how this system leverages federated multi-modal contrastive learning to enhance user authentication security while protecting privacy. Discover how FAMOS addresses critical challenges in mobile payment security by fusing multi-modal sensor data and clustering user data representations by action category, effectively eliminating background noise interference. Examine the framework's impressive performance metrics, including a 0.91 F1-Score and 0.97 AUC, representing significant improvements of 42.19% and 27.63% over baseline methods. Understand how the integration of federated learning not only maintains user privacy but also improves the system's overall effectiveness in real-world deployment scenarios.
FAMOS: Robust Privacy-Preserving Authentication on Payment Apps via Federated Multi-Modal Contrastive Learning
Overview
Syllabus
USENIX Security '24 - FAMOS: Robust Privacy-Preserving Authentication on Payment Apps via...
Taught by
USENIX
Related Courses
-
MD-ML: Super Fast Privacy-Preserving Machine Learning for Malicious Security with Dishonest Majority
-
Fundamentals of Privacy-Preserving Federated Learning
-
Secure Biometric Authentication Using Privacy-Preserving Cryptography
-
Nothing to Hide - Privacy-Preserving Cryptographic Authentication in Practice
-
Survey on Privacy and Security in Federated Learning
-
SecretFlow-SPU - A Performant and User-Friendly Framework for Privacy-Preserving Machine Learning
