From Heatmaps to Structural Explanations of Image Classifiers
Authors:
Li Fuxin,
Zhongang Qi,
Saeed Khorram,
Vivswan Shitole,
Prasad Tadepalli,
Minsuk Kahng,
Alan Fern
Abstract:
This paper summarizes our endeavors in the past few years in terms of explaining image classifiers, with the aim of including negative results and insights we have gained. The paper starts with describing the explainable neural network (XNN), which attempts to extract and visualize several high-level concepts purely from the deep network, without relying on human linguistic concepts. This helps us…
▽ More
This paper summarizes our endeavors in the past few years in terms of explaining image classifiers, with the aim of including negative results and insights we have gained. The paper starts with describing the explainable neural network (XNN), which attempts to extract and visualize several high-level concepts purely from the deep network, without relying on human linguistic concepts. This helps users understand network classifications that are less intuitive and substantially improves user performance on a difficult fine-grained classification task of discriminating among different species of seagulls.
Realizing that an important missing piece is a reliable heatmap visualization tool, we have developed I-GOS and iGOS++ utilizing integrated gradients to avoid local optima in heatmap generation, which improved the performance across all resolutions. During the development of those visualizations, we realized that for a significant number of images, the classifier has multiple different paths to reach a confident prediction. This has lead to our recent development of structured attention graphs (SAGs), an approach that utilizes beam search to locate multiple coarse heatmaps for a single image, and compactly visualizes a set of heatmaps by capturing how different combinations of image regions impact the confidence of a classifier.
Through the research process, we have learned much about insights in building deep network explanations, the existence and frequency of multiple explanations, and various tricks of the trade that make explanations work. In this paper, we attempt to share those insights and opinions with the readers with the hope that some of them will be informative for future researchers on explainable deep learning.
△ Less
Submitted 13 September, 2021;
originally announced September 2021.
One Explanation is Not Enough: Structured Attention Graphs for Image Classification
Authors:
Vivswan Shitole,
Li Fuxin,
Minsuk Kahng,
Prasad Tadepalli,
Alan Fern
Abstract:
Attention maps are a popular way of explaining the decisions of convolutional networks for image classification. Typically, for each image of interest, a single attention map is produced, which assigns weights to pixels based on their importance to the classification. A single attention map, however, provides an incomplete understanding since there are often many other maps that explain a classifi…
▽ More
Attention maps are a popular way of explaining the decisions of convolutional networks for image classification. Typically, for each image of interest, a single attention map is produced, which assigns weights to pixels based on their importance to the classification. A single attention map, however, provides an incomplete understanding since there are often many other maps that explain a classification equally well. In this paper, we introduce structured attention graphs (SAGs), which compactly represent sets of attention maps for an image by capturing how different combinations of image regions impact a classifier's confidence. We propose an approach to compute SAGs and a visualization for SAGs so that deeper insight can be gained into a classifier's decisions. We conduct a user study comparing the use of SAGs to traditional attention maps for answering counterfactual questions about image classifications. Our results show that the users are more correct when answering comparative counterfactual questions based on SAGs compared to the baselines.
△ Less
Submitted 7 November, 2021; v1 submitted 12 November, 2020;
originally announced November 2020.
