- Explainable Neural Networks: Recent Advancements, Part 1
- Explainable Neural Networks: Recent Advancements, Part 2
- Explainable Neural Networks: Recent Advancements, Part 3
- Network In Network by Min Lin 2013
Gradient Based Approaches
Visualizing Gradients (2014)
In the paper “Deep Inside Convolutional Networks: Visualising Image Classification Models and Saliency Maps“ Karen Simonyan et. al. proposed two visualisation techniques, based on computing the gradient of the class score with respect to the input image.
- The first technique generates an image, which maximises the class score, visualising the notion of the class, captured by a ConvNet.
- The second technique computes a class saliency map, specific to a given image and class.