Table 2.

Saturated performances of stochastic channel-based federated learning with pruning with different update rates.

Update rate	AUC-ROCa	AUC-PRb
10%	0.9776	0.9695
20%	0.9772	0.9686
30%	0.9777	0.9697
40%	0.9768	0.9604
50%	0.9780	0.9695
60%	0.9774	0.9682
70%	0.9774	0.9688
80%	0.9781	0.9703
90%	0.9774	0.9676
100%	0.9775	0.9685

^aAUC-ROC: area under the receiver operating characteristic curve.

^bAUC-PR: area under the precision-recall curve.