Axiom-based Grad-CAM: Towards Accurate Visualization and Explanation of CNNs
This work addresses the need for more theoretically grounded visualization methods in deep learning, particularly for researchers and practitioners using CNNs, though it is incremental as it builds on existing CAM approaches.
The paper tackled the problem of visualizing and explaining Convolutional Neural Networks (CNNs) by introducing two axioms—Conservation and Sensitivity—to improve Class Activation Mapping (CAM) methods, resulting in XGrad-CAM, which achieved better visualization performance than Grad-CAM while being class-discriminative and easy-to-implement.
To have a better understanding and usage of Convolution Neural Networks (CNNs), the visualization and interpretation of CNNs has attracted increasing attention in recent years. In particular, several Class Activation Mapping (CAM) methods have been proposed to discover the connection between CNN's decision and image regions. In spite of the reasonable visualization, lack of clear and sufficient theoretical support is the main limitation of these methods. In this paper, we introduce two axioms -- Conservation and Sensitivity -- to the visualization paradigm of the CAM methods. Meanwhile, a dedicated Axiom-based Grad-CAM (XGrad-CAM) is proposed to satisfy these axioms as much as possible. Experiments demonstrate that XGrad-CAM is an enhanced version of Grad-CAM in terms of conservation and sensitivity. It is able to achieve better visualization performance than Grad-CAM, while also be class-discriminative and easy-to-implement compared with Grad-CAM++ and Ablation-CAM. The code is available at https://github.com/Fu0511/XGrad-CAM.