the result between the actual and desired values does not match.
The reason for this error is in the learning process. Error in learning
is because of three main factors; bias, variance and noise [7].
Error ¼ Noise þ Bias þ Variance ð1Þ
Large bias causes under-fitting while large variance causes
over-fitting of data. Generalization of an ANN to adapt to the
new data set will increase much if bias and variance are reduced
to minimum as data will not suffer the problem of over-fitting
and under-fitting anymore. Moreover, the error in learning also
reduces and overall prediction improves.