# Image-Specific Class Saliency Visualisation

In the paper "Deep Inside Convolutional Networks: Visualising Image Classification Models and Saliency Maps", https://arxiv.org/abs/1312.6034, at part 3, there is a first-order Taylor expansion(formula number 3 & 4) that I can't understand the logic behind it and how they are obtained.

Those formulas are about computing saliency map in convolutional neural networks.