Federated learning has been widely applied to enable decentralized devices, which each have their own local data, to learn a shared model. However, learning from real-world data can be challenging, as it is rarely identically and independently distributed (IID) across edge devices (a key assumption for current high-performing and low-bandwidth algorithms). We present a novel approach, FedCD, which clones and deletes models to dynamically group devices with similar data. Experiments on the CIFAR-10 dataset show that FedCD achieves higher accuracy and faster convergence compared to a FedAvg baseline on non-IID data while incurring minimal computation, communication, and storage overheads.
@article{arxiv.2006.09637,
title = {FedCD: Improving Performance in non-IID Federated Learning},
author = {Kavya Kopparapu and Eric Lin and Jessica Zhao},
journal= {arXiv preprint arXiv:2006.09637},
year = {2020}
}
Comments
Accepted for Oral Presentation at ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (KDD 2020) International workshop on Artificial Intelligence of Things