FedUNet: A lightweight additive U-Net module for federated learning with heterogeneous models
K. Lee, Beomseok Seo, J. Park · ICT Express · 2026
Federated learning (FL) enables decentralized model training without sharing data, but most existing methods assume homogeneous model architectures across clients, limiting applicability in heterogeneous real-world environments. We propose FedUNet , a lightweight and architecture-agnostic FL framework that introduces an additive U-Net-inspired module attached to each client’s backbone. By sharing only the compact bottleneck of the U-Net, FedUNet enables efficient knowledge transfer without architectural alignment, shared data, or costly distillation. The encoder–decoder structure with skip connections integrates multi-scale features into a shared latent space, promoting client-invariant representation alignment across heterogeneous models. Experimental results on CIFAR-10 and CIFAR-100 with VGG and ResNet backbones show that FedUNet achieves high accuracy with significantly reduced communication and computational overhead. Notably, a full U-Net attains 93.92% accuracy while requiring only 0.89 MB of communication per round, highlighting its practicality for resource-constrained FL systems.