Generating Synthetic Data in a Secure Federated General Adversarial Networks for a Consortium of Health Registries

Abstract

In this work, we review the architecture design of existing federated General Adversarial Networks (GAN) solutions and highlight the security and trust-related weaknesses in the existing designs. We then describe how these weaknesses make existing designs unsuitable for the requirements needed for a consortium of health registries working towards generating synthetic datasets for research purposes. Moreover, we propose how these weaknesses can be addressed with our novel architecture solution. Our novel architecture solution combines several building blocks to generate synthetic data in a decentralised setting. Consortium blockchains, secure multi-party computations, and homomorphic encryption are the core building blocks of our proposed architecture solution to address the weaknesses in the existing design of federated GANs. Finally, we discuss our proposed solution's advantages and future research directions.