Fig. 5From: Federated learning algorithms for generalized mixed-effects model (GLMM) on horizontally partitioned data from distributed sourcesThe accuracy in 8 different data settings among centralized, Laplace, and Gauss–Hermite method under significance level \(\alpha =0.05\). The performance of models in different number of nodes (2 vs 10), sample size (500 vs 30), and variance (large vs small)Back to article page