SOURCE-LINKED INTELLIGENCE

Personalized Federated Learning through Global Knowledge Distillation and Local Head Adaptation

arXiv · AI, language, vision and robotics · article · Sep 15, 2026 · UTC

Statistical heterogeneity limits federated learning when a single global classifier cannot represent client-specific label distributions. In this work, we propose Personalized Federated Knowledge Distillation with Head Adaptation (pFedKDH), which aggregates only the shared backbone, keeps persistent client-specific heads, and uses a recalibrated global head as a teacher during local training. Across MNIST, Fashion-MNIST, CIFAR10, and CIFAR100 under class-wise Dirichlet partitions, pFedKDH obtains the best accuracy in most settings, with accuracy gaps up to 37.67\% over the weakest baseline and

Read original source ↗ Open in workspace

recordType

paper

region

Global