Optimisation of federated learning settings under statistical heterogeneity variations
This work addresses the challenge of statistical heterogeneity in federated learning for practitioners, but it is incremental as it builds on existing FL methods with empirical optimizations.
The paper tackled the problem of optimizing federated learning (FL) performance under varying levels of statistical heterogeneity by conducting an empirical analysis of FL parameters and aggregators across three datasets, resulting in recommended guidelines for parameter selection to improve model performance.
Federated Learning (FL) enables local devices to collaboratively learn a shared predictive model by only periodically sharing model parameters with a central aggregator. However, FL can be disadvantaged by statistical heterogeneity produced by the diversity in each local devices data distribution, which creates different levels of Independent and Identically Distributed (IID) data. Furthermore, this can be more complex when optimising different combinations of FL parameters and choosing optimal aggregation. In this paper, we present an empirical analysis of different FL training parameters and aggregators over various levels of statistical heterogeneity on three datasets. We propose a systematic data partition strategy to simulate different levels of statistical heterogeneity and a metric to measure the level of IID. Additionally, we empirically identify the best FL model and key parameters for datasets of different characteristics. On the basis of these, we present recommended guidelines for FL parameters and aggregators to optimise model performance under different levels of IID and with different datasets