Loosely Coupled Federated Learning Over Generative Models

Abstract

Federated learning (FL) was proposed to achieve collaborative machine learning among various clients without uploading private data. However, due to model aggregation strategies, existing frameworks require strict model homogeneity, limiting the application in more complicated scenarios. Besides, the communication cost of FL's model and gradient transmission is extremely high. This paper proposes Loosely Coupled Federated Learning (LC-FL), a framework using generative models as transmission media to achieve low communication cost and heterogeneous federated learning. LC-FL can be applied on scenarios where clients possess different kinds of machine learning models. Experiments on real-world datasets covering different multiparty scenarios demonstrate the effectiveness of our proposal.

References in corpus (6)

• Fashion-MNIST: a Novel Image Dataset for Benchmarking Machine Learning Algorithms
• Private federated learning on vertically partitioned data via entity resolution and additively homomorphic encryption
• Variational Autoencoder for Deep Learning of Images, Labels and Captions
• Learning to Protect Communications with Adversarial Neural Cryptography
• Generative Models for Effective ML on Private, Decentralized Datasets
• GAN and VAE from an Optimal Transport Point of View