The field of machine learning is moving towards a more secure and privacy-preserving direction, with a focus on federated learning, homomorphic encryption, and secure multiparty computation. Recent developments have shown that it is possible to achieve high accuracy and security in machine learning models while protecting sensitive data. The use of federated learning, in particular, has gained significant attention in recent years, as it enables collaborative model training without centralizing client data. Noteworthy papers in this area include FuSeFL, which presents a fully secure and scalable federated learning scheme designed for cross-silo settings, and VMask, which proposes a novel label privacy protection framework for vertical federated learning via layer masking. These papers demonstrate the potential for machine learning to be both secure and accurate, and highlight the importance of continued research in this area.
Advances in Secure and Privacy-Preserving Machine Learning
Sources
FuSeFL: Fully Secure and Scalable Cross-Silo Federated Learning
Towards Efficient Privacy-Preserving Machine Learning: A Systematic Review from Protocol, Model, and System Perspectives
VTarbel: Targeted Label Attack with Minimal Knowledge on Detector-enhanced Vertical Federated Learning
VMask: Tunable Label Privacy Protection for Vertical Federated Learning via Layer Masking
Privacy-Preserving Drone Navigation Through Homomorphic Encryption for Collision Avoidance
Collusion-Resilient Hierarchical Secure Aggregation with Heterogeneous Security Constraints
Careful Whisper: Attestation for peer-to-peer Confidential Computing networks
A Privacy-Centric Approach: Scalable and Secure Federated Learning Enabled by Hybrid Homomorphic Encryption
AnalogFed: Federated Discovery of Analog Circuit Topologies with Generative AI
Privacy-Preserving Multimodal News Recommendation through Federated Learning
FedMultiEmo: Real-Time Emotion Recognition via Multimodal Federated Learning
Challenges of Trustworthy Federated Learning: What's Done, Current Trends and Remaining Work
Verifying International Agreements on AI: Six Layers of Verification for Rules on Large-Scale AI Development and Deployment
Federated Learning for Large-Scale Cloud Robotic Manipulation: Opportunities and Challenges