Generative AI-Powered Plugin for Robust Federated Learning in Heterogeneous IoT Networks
This addresses data heterogeneity issues in federated learning for IoT networks, offering an incremental enhancement to existing optimization techniques.
The paper tackles the problem of Non-IID data distribution hindering convergence and performance in federated learning for IoT networks by proposing a plugin that uses generative AI for data augmentation and balanced sampling, resulting in significant improvements in convergence speed and robustness against data imbalance.
Federated learning enables edge devices to collaboratively train a global model while maintaining data privacy by keeping data localized. However, the Non-IID nature of data distribution across devices often hinders model convergence and reduces performance. In this paper, we propose a novel plugin for federated optimization techniques that approximates Non-IID data distributions to IID through generative AI-enhanced data augmentation and balanced sampling strategy. Key idea is to synthesize additional data for underrepresented classes on each edge device, leveraging generative AI to create a more balanced dataset across the FL network. Additionally, a balanced sampling approach at the central server selectively includes only the most IID-like devices, accelerating convergence while maximizing the global model's performance. Experimental results validate that our approach significantly improves convergence speed and robustness against data imbalance, establishing a flexible, privacy-preserving FL plugin that is applicable even in data-scarce environments.