Improve federated continual learning with FEAT: geometry-aware correction enhances exemplar replay under dynamic data heterogeneity and class imbalance.
Discover SubFLOT, a novel framework enhancing federated learning efficiency and personalization using optimal transport and adaptive regularization techniq...
Enhance medical image segmentation with adaptive differential privacy in federated learning, ensuring privacy and high accuracy across diverse imaging moda...
Improve federated learning accuracy and security using server learning, client update filtering, and geometric median aggregation against malicious attacks...