MULTI-FLGANs: Multi-Distributed Adversarial Networks for Non-IID distribution
This addresses the challenge of training GANs in federated learning settings with non-IID data, which is incremental as it builds on existing federated GAN architectures.
The paper tackled the problem of instability and mode collapse in federated GANs for non-IID datasets by proposing MULTI-FLGAN, resulting in a fourfold increase in stability and performance on average over 20 clients compared to baseline FLGAN.
Federated learning is an emerging concept in the domain of distributed machine learning. This concept has enabled GANs to benefit from the rich distributed training data while preserving privacy. However, in a non-iid setting, current federated GAN architectures are unstable, struggling to learn the distinct features and vulnerable to mode collapse. In this paper, we propose a novel architecture MULTI-FLGAN to solve the problem of low-quality images, mode collapse and instability for non-iid datasets. Our results show that MULTI-FLGAN is four times as stable and performant (i.e. high inception score) on average over 20 clients compared to baseline FLGAN.