Tag: non-IID data

A critical review analyzes privacy-preserving federated learning developments published between 2024 and early 2026 across healthcare applications, examining technical hurdles ...

A hierarchical federated learning architecture integrating semi-asynchronous client uploads and scheduled cloud synchronization cut training time by 52 percent while ...

Researchers demonstrated that the FedDefend++ framework reinforces federated learning against model inversion and data poisoning attacks through trust-weighted aggregation, while ...

A new federated learning framework reads each school's own CO2 record to set its aggregation weight, matching centralized training on ...

A new study shows federated learning's average accuracy can mask severe failures for individual clients, and that server-side re-weighting destabilizes ...

A new decentralized federated learning framework uses particle swarm optimization to dynamically select clients, reporting 76.2% accuracy on CIFAR-10 under ...

FedSPA combines Shapley-value analysis and model parameter similarity to identify malicious clients in federated learning under data poisoning attacks.

Researchers propose CROG and CLIG, two aggregation algorithms that weight client updates by Euclidean gradient distances to stabilize federated learning ...

A new personalized federated learning framework called FedACC dynamically separates shared and personalized model parameters and clusters clients, reporting the ...

A new federated learning framework for chest X-ray analysis reports modest accuracy gains over existing methods while aligning explanations across ...