An experimental study of class imbalance in federated learning

Federated learning is a distributed machine learning paradigm that trains a global model for prediction based on several local models at clients while local data privacy is preserved. Class imbalance is believed to be one of the factors that degrades the global model performance. However, there has been very little research on if and how class imbalance can affect the global performance in various imbalance scenarios. Class imbalance in federated learning is much more complex than that in traditional non-distributed machine learning, due to different class imbalance situations at local clients. Class imbalance needs to be re-defined in distributed learning environments, so that corresponding solutions can be proposed. In this paper, first, we propose two new metrics to define class imbalance – the global class imbalance degree (MID) and the local difference of class imbalance among clients (WCS). Class imbalance is categorized into four scenarios under the definition. Then, we conduct extensive experiments to analyze the impact of class imbalance on the global performance in various scenarios. Our results show that a higher MID and a larger WCS degrade more the performance of the global model. Besides, WCS is shown to slow down the convergence of the global model by misdirecting the optimization.