Delta-DAGMM: A Free Rider Attack Detection Model in Horizontal Federated Learning

Table 3

Precision of Delta-DAGMM in different strategies of Free-rider attack detection.

Dataset	Type	Precision (%)
		Plain (i)	Plain (ii)	Plain (iii)	Disguised (i)	Disguised (ii)
MLP-Federate	Single time	98.2	95.7	100	83.9	83.3
	Overall	100	99.8	100	96.6	96.0
CNN-Federate	Single time	98.4	95.2	100	82.9	83.2
	Overall	100	99.6	100	95.8	96.4