Neural network disruption caused by inaccuracies machines, and they can be simulated by embedding the original input or interference with the connection weights, allowing us to study the characteristics of functions under small disturbance parameter. These include sensitivity analysis of multilayer perceptron neural network and radial basis function neural networks, two models are widely used in the field of machine learning. This book will be useful to the engineer to apply the sensitivity analysis of neural networks to solve practical problems, and for researchers interested in fundamental problems in neural networks.