Visual Explanations from Deep Networks via Riemann-Stieltjes Integrated Gradient-based Localization
This addresses the need for more reliable and interpretable visual explanations in deep learning for image classification, though it is incremental as it builds on existing methods.
The paper tackles the problem of generating visual explanations for CNN predictions by introducing a new technique that combines the layer-wise applicability of Grad-CAM with the gradient stability of Integrated Gradients, using a Riemann-Stieltjes sum approximation for efficiency, resulting in heatmaps that are better focused and more stable than Grad-CAM.
Neural networks are becoming increasingly better at tasks that involve classifying and recognizing images. At the same time techniques intended to explain the network output have been proposed. One such technique is the Gradient-based Class Activation Map (Grad-CAM), which is able to locate features of an input image at various levels of a convolutional neural network (CNN), but is sensitive to the vanishing gradients problem. There are techniques such as Integrated Gradients (IG), that are not affected by that problem, but its use is limited to the input layer of a network. Here we introduce a new technique to produce visual explanations for the predictions of a CNN. Like Grad-CAM, our method can be applied to any layer of the network, and like Integrated Gradients it is not affected by the problem of vanishing gradients. For efficiency, gradient integration is performed numerically at the layer level using a Riemann-Stieltjes sum approximation. Compared to Grad-CAM, heatmaps produced by our algorithm are better focused in the areas of interest, and their numerical computation is more stable. Our code is available at https://github.com/mlerma54/RSIGradCAM