Calibrated One Round Federated Learning with Bayesian Inference in the Predictive Space
This addresses the need for well-calibrated models in federated learning with small, noisy, and heterogeneous data, representing an incremental improvement over existing Bayesian FL methods.
The paper tackled the problem of overconfident predictions in federated learning by proposing β-Predictive Bayes, a Bayesian FL algorithm that interpolates between mixture and product of predictive posteriors to improve calibration, showing superiority in calibration on various datasets as data heterogeneity increases.
Federated Learning (FL) involves training a model over a dataset distributed among clients, with the constraint that each client's dataset is localized and possibly heterogeneous. In FL, small and noisy datasets are common, highlighting the need for well-calibrated models that represent the uncertainty of predictions. The closest FL techniques to achieving such goals are the Bayesian FL methods which collect parameter samples from local posteriors, and aggregate them to approximate the global posterior. To improve scalability for larger models, one common Bayesian approach is to approximate the global predictive posterior by multiplying local predictive posteriors. In this work, we demonstrate that this method gives systematically overconfident predictions, and we remedy this by proposing $β$-Predictive Bayes, a Bayesian FL algorithm that interpolates between a mixture and product of the predictive posteriors, using a tunable parameter $β$. This parameter is tuned to improve the global ensemble's calibration, before it is distilled to a single model. Our method is evaluated on a variety of regression and classification datasets to demonstrate its superiority in calibration to other baselines, even as data heterogeneity increases. Code available at https://github.com/hasanmohsin/betaPredBayesFL