Grad-CAM: Why did you say that?

Abstract

We propose a technique for making Convolutional Neural Network (CNN)-based models more transparent by visualizing input regions that are 'important' for predictions -- or visual explanations. Our approach, called Gradient-weighted Class Activation Mapping (Grad-CAM), uses class-specific gradient information to localize important regions. These localizations are combined with existing pixel-space visualizations to create a novel high-resolution and class-discriminative visualization called Guided Grad-CAM. These methods help better understand CNN-based models, including image captioning and visual question answering (VQA) models. We evaluate our visual explanations by measuring their ability to discriminate between classes, to inspire trust in humans, and their correlation with occlusion maps. Grad-CAM provides a new way to understand CNN-based models. We have released code, an online demo hosted on CloudCV, and a full version of this extended abstract.

Presented at NIPS 2016 Workshop on Interpretable Machine Learning in Complex Systems. This is an extended abstract version of arXiv:1610.02391 (CVPR format)

References in corpus (1)

• Striving for Simplicity: The All Convolutional Net

Cited by in corpus (13)

• Demystifying Contrastive Self-Supervised Learning: Invariances, Augmentations and Dataset Biases
• Do Explanations Reflect Decisions? A Machine-centric Strategy to Quantify the Performance of Explainability Algorithms
• A deep convolutional neural network for classification of Aedes albopictus mosquitoes
• Neural Network Attributions: A Causal Perspective
• Bias-based Universal Adversarial Patch Attack for Automatic Check-out
• Explainable Semantic Federated Learning Enabled Industrial Edge Network for Fire Surveillance
• A Simple Framework for Uncertainty in Contrastive Learning
• Using KL-divergence to focus Deep Visual Explanation
• SCALP -- Supervised Contrastive Learning for Cardiopulmonary Disease Classification and Localization in Chest X-rays using Patient Metadata
• Survival-oriented embeddings for improving accessibility to complex data structures
• Deep Learning for Wireless Communications
• Explainable Deep Reinforcement Learning for Portfolio Management: An Empirical Approach
• e-QRAQ: A Multi-turn Reasoning Dataset and Simulator with Explanations