Understanding Integrated Gradients with SmoothTaylor for Deep Neural Network Attribution
- 10 January 2021
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 4949-4956
Abstract
Integrated Gradients as an attribution method for deep neural network models offers simple implementability. However, it suffers from noisiness of explanations which affects the ease of interpretability. The SmoothGrad technique is proposed to solve the noisiness issue and smoothen the attribution maps of any gradient-based attribution method. In this paper, we present SmoothTaylor as a novel theoretical concept bridging Integrated Gradients and SmoothGrad, from the Taylor's theorem perspective. We apply the methods to the image classification problem, using the ILSVRC2012 ImageNet object recognition dataset, and a couple of pretrained image models to generate attribution maps. These attribution maps are empirically evaluated using quantitative measures for sensitivity and noise level. We further propose adaptive noising to optimize for the noise scale hyperparameter value. From our experiments, we find that the SmoothTaylor approach together with adaptive noising is able to generate better quality saliency maps with lesser noise and higher sensitivity to the relevant points in the input space as compared to Integrated Gradients.Keywords
All Related Versions
This publication has 13 references indexed in Scilit:
- Densely Connected Convolutional Networks2017
- Universal Adversarial Perturbations2017
- Explaining nonlinear classification decisions with deep Taylor decompositionPattern Recognition, 2017
- Evaluating the Visualization of What a Deep Neural Network Has LearnedIEEE Transactions on Neural Networks and Learning Systems, 2016
- Layer-Wise Relevance Propagation for Neural Networks with Local Renormalization Layers2016
- Faster R-CNN: Towards Real-Time Object Detection with Region Proposal NetworksIEEE Transactions on Pattern Analysis and Machine Intelligence, 2016
- Learning Deep Features for Discriminative Localization2016
- Deep neural networks are easily fooled: High confidence predictions for unrecognizable images2015
- ImageNet Large Scale Visual Recognition ChallengeInternational Journal of Computer Vision, 2015
- The Laplacian Pyramid as a Compact Image CodeIEEE Transactions on Communications, 1983