Fig. 3From: Federated learning algorithms for generalized mixed-effects model (GLMM) on horizontally partitioned data from distributed sourcesThe comparison of performances between centralized GLM and federated GH models. (Left) The centralized GLM with Logistic Regression; (Right) The federated GLMM with 2-degree GH approximation. All methods are applied to all 8 settings of synthetic data that pooled together. Method “centralized GLM”is done with R packages “lm” in a centralized settingsBack to article page