Cybersecurity

Table 1 Comparison of clone accuracy rates and extraction rates obtained from different methods

From: FMSA: a meta-learning framework-based fast model stealing attack technique against intelligent network intrusion detection systems

Dataset	Victim ACC (%)	Attacks	Clone ACC (%)	MER (%)
CICIDS2017	96.71	FMSA	88.92	90.10
		Random	48.64	41.39
		Knockoff	69.15	57.10
		MAZE	80.83	82.50
MCFP	92.36	FMSA	83.85	87.74
		Random	37.28	22.75
		Knockoff	62.73	51.46
		MAZE	75.48	79.47

Back to article page