The field of federated learning and secure data exchange is rapidly evolving, with a focus on developing innovative solutions to address challenges such as data privacy, security, and personalization. Recent research has explored the use of novel techniques, including balanced batch normalization, style-aware transformer aggregation, and hierarchical secure aggregation, to improve the robustness and accuracy of federated learning models. Additionally, there is a growing interest in developing secure data exchange mechanisms, such as decentralized data exchange and homomorphic encryption, to enable the secure sharing of data across different entities. Noteworthy papers in this area include pFedBBN, which proposes a personalized federated test-time adaptation framework, and DEXO, which introduces a secure and fair exchange mechanism for decentralized IoT data markets. Furthermore, research on federated vision transformer learning and trustless federated learning at edge-scale is also making significant progress, enabling the development of more robust and secure machine learning models.
Advancements in Federated Learning and Secure Data Exchange
Sources
pFedBBN: A Personalized Federated Test-Time Adaptation with Balanced Batch Normalization for Class-Imbalanced Data
ioPUF+: A PUF Based on I/O Pull-Up/Down Resistors for Secret Key Generation in IoT Nodes
Real-Time Personalized Content Adaptation through Matrix Factorization and Context-Aware Federated Learning
DEXO: A Secure and Fair Exchange Mechanism for Decentralized IoT Data Markets
Federated style aware transformer aggregation of representations
Personalized Federated Segmentation with Shared Feature Aggregation and Boundary-Focused Calibration
Hi-SAFE: Hierarchical Secure Aggregation for Lightweight Federated Learning
Peregrine: One-Shot Fine-Tuning for FHE Inference of General Deep CNNs
FedPoisonTTP: A Threat Model and Poisoning Attack for Federated Test-Time Personalization
DISCO: A Browser-Based Privacy-Preserving Framework for Distributed Collaborative Learning
On hierarchical secure aggregation against relay and user collusion
Privacy-Preserving Federated Vision Transformer Learning Leveraging Lightweight Homomorphic Encryption in Medical AI
Trustless Federated Learning at Edge-Scale: A Compositional Architecture for Decentralized, Verifiable, and Incentive-Aligned Coordination