Fig. 2From: Federated learning algorithms for generalized mixed-effects model (GLMM) on horizontally partitioned data from distributed sourcesThe difference from coefficients to the true parameters that are used to generate data. (Left) The distributed GLMM with Laplace approximation; (Middle) The distributed GLMM with 2-degree Gauss–Hermite approximation. Reminds that \(X_1\) is the intercept; (Right) The benchmark of centralized GLMM in R packageBack to article page