Abstract
A linearly separable Boolean function is derived from a set of examples by a perceptron with optimal stability. The probability to reconstruct a pattern which is not learnt is calculated analytically using the replica method.
Click here to close this panel.
M Opper1, W Kinzel1, J Kleinz1 and R Nehl1
Published under licence by IOP Publishing Ltd
,
,
Citation M Opper et al 1990 J. Phys. A: Math. Gen. 23 L581
DOI 10.1088/0305-4470/23/11/012
392 Total downloads
A linearly separable Boolean function is derived from a set of examples by a perceptron with optimal stability. The probability to reconstruct a pattern which is not learnt is calculated analytically using the replica method.