An Adaptive Communication-Efficient Federated Learning to Resist Gradient-Based Reconstruction Attacks

Table 6

The comparative experiment results.

Function			The parameters of
AFC-FL	570,434	86	5.265 × 10^7	0.9597
F = 1, CE = 100	620,060	98	6.076 × 10^7	0.9738
F = 1, CE = 90	558,042	329	1.836 × 10^8	0.9166
F = 5, CE = 100	620,060	221	1.370 × 10^8	0.9622
F = 5, CE = 90	558,039	459	2.561 × 10^8	0.9092