whether saliency methods are insensitive to model and data
Insensitivity is highly undesirable, because it would mean that the “explanation” isunrelated to model and data
Methods that are insensitive to model and training data are similar to edge detectors
Edge detectors simply highlight strong pixel color changes in images and areunrelated to a prediction model or abstract features of the image, and require no training
The methods tested were Vanilla Gradient, Gradient x Input, Integrated Gradients,Guided Backpropagation, Guided Grad-CAM and SmoothGrad (with VanillaGradient).
Vanilla Gradient and Grad-CAM passed the insensitivity check, while GuidedBackpropagation and Guided Grad-CAM failed
However, the sanity checks paper itself has found some criticism from Tomsett et al.(2020) 89 with a paper called “Sanity checks for caliency metrics” (of course
They found that there is a lack of consistency for evaluation metrics
So we are back to where we started … It remains difficult to evaluate the visual explanations. This makes it very difficult for a practitioner.