VisualBackProp: efficient visualization of CNNs

Abstract

This paper proposes a new method, that we call VisualBackProp, for visualizing which sets of pixels of the input image contribute most to the predictions made by the convolutional neural network (CNN). The method heavily hinges on exploring the intuition that the feature maps contain less and less irrelevant information to the prediction decision when moving deeper into the network. The technique we propose was developed as a debugging tool for CNN-based systems for steering self-driving cars and is therefore required to run in real-time, i.e. it was designed to require less computations than a forward propagation. This makes the presented visualization method a valuable debugging tool which can be easily used during both training and inference. We furthermore justify our approach with theoretical arguments and theoretically confirm that the proposed method identifies sets of input pixels, rather than individual pixels, that collaboratively contribute to the prediction. Our theoretical findings stand in agreement with the experimental results. The empirical evaluation shows the plausibility of the proposed approach on the road video data as well as in other applications and reveals that it compares favorably to the layer-wise relevance propagation approach, i.e. it obtains similar visualization results and simultaneously achieves order of magnitude speed-ups.

References in corpus (7)

• Explaining and Harnessing Adversarial Examples
• Intriguing properties of neural networks
• End to End Learning for Self-Driving Cars
• Understanding Neural Networks Through Deep Visualization
• Visualizing Deep Neural Network Decisions: Prediction Difference Analysis
• DeepDriving: Learning Affordance for Direct Perception in Autonomous Driving
• Evaluating the visualization of what a Deep Neural Network has learned

Cited by in corpus (12)

• An Explainable 3D Residual Self-Attention Deep Neural Network FOR Joint Atrophy Localization and Alzheimer's Disease Diagnosis using Structural MRI
• Quantum Algorithms for Deep Convolutional Neural Networks
• A Survey of Deep Network Solutions for Learning Control in Robotics: From Reinforcement to Imitation
• A Survey of Deep Learning for Scientific Discovery
• End-to-end Multi-Modal Multi-Task Vehicle Control for Self-Driving Cars with Visual Perception
• Rethinking Self-driving: Multi-task Knowledge for Better Generalization and Accident Explanation Ability
• Causal Navigation by Continuous-time Neural Networks
• Predicting Model Failure using Saliency Maps in Autonomous Driving Systems
• Towards Cooperative Flight Control Using Visual-Attention
• Deep Relevance Regularization: Interpretable and Robust Tumor Typing of Imaging Mass Spectrometry Data
• Quantum Algorithms for Unsupervised Machine Learning and Neural Networks
• Imitation Learning via Simultaneous Optimization of Policies and Auxiliary Trajectories