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

Table 13

Minterm-based decomposition in Example 6.

	Template values	Logic operator	Margin
1	𝑤=(−5,−2,−1,1,5,−3,0,0,0),𝑏=−8		0.1240
2	𝑤=(3,1,−3,3,−1,1,0,0,0),𝑏=−7	OR	0.1826
3	𝑤=(3,−2,1,−3,1,−1,0,0,0),𝑏=−6	OR	0.2000
4	𝑤=(−2,2,1,1,−2,−2,0,0,0),𝑏=−7	OR	0.2357
5	𝑤=(1,1,1,1,0,1,0,0,0),𝑏=−4	OR	0.4472