Using the Taylor expansion of multilayer feedforward neural networks
Andries Petrus Engelbrecht · 2000
The Taylor series expansion of continuous functions has shown - in many fields - to be an extremely powerful tool to study the characteristics of such functions. This paper illustrates the power of the Taylor series expansion of multilayer feedforward neural networks. The paper shows how these expansions can be used to investigate positions of decision boundaries, to develop active learning strategies and to perform architecture selection.