Robust Template Decomposition without Weight Restriction for Cellular Neural Networks Implementing Arbitrary Boolean Functions Using Support Vector Classifiers

Table 6

LHJ decomposition in Example 3.

	Template values	Logic operator	Margin
1	𝑤=(3,2,2,2,3,1,2,1,2,1),𝑏=1		0.1581
2	𝑤=(1,1,2,1,2,0,1,1,1),𝑏=1	AND	0.2673
3	𝑤=(2,1,0,1,0,1,1,0,1),𝑏=−4	OR	0.3333
4	𝑤=(0,1,2,1,2,0,1,1,1),𝑏=−2	OR	0.3333
5	𝑤=(1,1,0,0,0,1,0,1,1),𝑏=4	AND	0.4472