Saliency methods are frequently used to explain Deep Neural Network-based models. Adebayo et al.'s work on evaluating saliency methods for classification models illustrate certain explanation methods fail the model and data randomization tests. However, on extending the tests for various state of the art object detectors we illustrate that the ability to explain a model is more dependent on the model itself than the explanation method. We perform sanity checks for object detection and define new qualitative criteria to evaluate the saliency explanations, both for object classification and bounding box decisions, using Guided Backpropagation, Integrated Gradients, and their Smoothgrad versions, together with Faster R-CNN, SSD, and EfficientDet-D0, trained on COCO. In addition, the sensitivity of the explanation method to model parameters and data labels varies class-wise motivating to perform the sanity checks for each class. We find that EfficientDet-D0 is the most interpretable method independent of the saliency method, which passes the sanity checks with little problems.
Backpropagation image saliency aims at explaining model predictions by estimating model-centric importance of individual pixels in the input. However, class-insensitivity of the earlier layers in a network only allows saliency computation with low resolution activation maps of the deeper layers, resulting in compromised image saliency. Remedifying this can lead to sanity failures. We propose CAMERAS, a technique to compute high-fidelity backpropagation saliency maps without requiring any external priors and preserving the map sanity. Our method systematically performs multi-scale accumulation …
Weizmann Institute of Science（魏茨曼科学研究学院）；Jether Energy Research（Jether能源研究院）
Saliency methods are a popular approach for model debugging and explainability. However, in the absence of ground-truth data for what the correct maps should be, evaluating and comparing different approaches remains a long-standing challenge. The sanity checks methodology of Adebayo et al [Neurips 2018] has sought to address this challenge. They argue that some popular saliency methods should not be used for explainability purposes since the maps they produce are not sensitive to the underlying model that is to be explained. Through a causal re-framing of their objective, we argue that their e…
IBM Research（IBM研究院）；Cardiff University（卡迪夫大学）；Crime and Security Research Institute（犯罪与安全研究所）；Booz Allen Hamilton（博思艾伦咨询公司）；CCDC Army Research Laboratory（CCDC陆军研究实验室）
Saliency maps are a popular approach to creating post-hoc explanations of image classifier outputs. These methods produce estimates of the relevance of each pixel to the classification output score, which can be displayed as a saliency map that highlights important pixels. Despite a proliferation of such methods, little effort has been made to quantify how good these saliency maps are at capturing the true relevance of the pixels to the classifier output (i.e. their "fidelity"). We therefore investigate existing metrics for evaluating the fidelity of saliency methods (i.e. saliency metrics). W…
Google Brain（谷歌大脑）；University of California Berkeley（加州大学伯克利分校）
Saliency methods have emerged as a popular tool to highlight features in an input deemed relevant for the prediction of a learned model. Several saliency methods have been proposed, often guided by visual appeal on image data. In this work, we propose an actionable methodology to evaluate what kinds of explanations a given method can and cannot provide. We find that reliance, solely, on visual assessment can be misleading. Through extensive experiments we show that some existing saliency methods are independent both of the model and of the data generating process. Consequently, methods that fa…
TU Berlin（柏林工业大学）；Fraunhofer HHI（弗劳恩霍夫海因里希·赫兹研究所）；Leibniz Institute of Agricultural Engineering and Bioeconomy e.V. (ATB)（莱布尼茨农业工程与生物经济研究所）；BIFOLD – Berlin Institute for the Foundations of Learning and Data（柏林学习与数据基础研究所）；University of Potsdam（波茨坦大学）
The University of Munich（慕尼黑大学）；Siemens AG（西门子股份公司）
The classification decisions of neural networks can be misled by small imperceptible perturbations. This work aims to explain the misled classifications using saliency methods. The idea behind saliency methods is to explain the classification decisions of neural networks by creating so-called saliency maps. Unfortunately, a number of recent publications have shown that many of the proposed saliency methods do not provide insightful explanations. A prominent example is Guided Backpropagation (GuidedBP), which simply performs (partial) image recovery. However, our numerical analysis shows the sa…
Indian Institute of Technology, Kharagpur（印度理工学院克勒格布尔分校）
Deep Learning has become overly complicated and has enjoyed stellar success in solving several classical problems like image classification, object detection, etc. Several methods for explaining these decisions have been proposed. Black-box methods to generate saliency maps are particularly interesting due to the fact that they do not utilize the internals of the model to explain the decision. Most black-box methods perturb the input and observe the changes in the output. We formulate saliency map generation as a sequential search problem and leverage upon Reinforcement Learning (RL) to accumu…
Our overarching goal is to develop an accurate and explainable model for plant disease identification using hyperspectral data. Charcoal rot is a soil borne fungal disease that affects the yield of soybean crops worldwide. Hyperspectral images were captured at 240 different wavelengths in the range of 383 - 1032 nm. We developed a 3D Convolutional Neural Network model for soybean charcoal rot disease identification. Our model has classification accuracy of 95.73\% and an infected class F1 score of 0.87. We infer the trained model using saliency map and visualize the most sensitive pixel locati…
University of Wisconsin - Milwaukee（威斯康星大学密尔沃基分校）；University of Southern California（南加州大学）
Predicting where people look in natural scenes has attracted a lot of interest in computer vision and computational neuroscience over the past two decades. Two seemingly contrasting categories of cues have been proposed to influence where people look: \textit{low-level image saliency} and \textit{high-level semantic information}. Our first contribution is to take a detailed look at these cues to confirm the hypothesis proposed by Henderson~\cite{henderson1993eye} and Nuthmann \& Henderson~\cite{nuthmann2010object} that observers tend to look at the center of objects. We analyzed fixation data …
This document summarizes different visual explanations methods such as CAM, Grad-CAM, Localization using Multiple Instance Learning - Saliency-based methods, Saliency-driven Class-Impressions, Muting pixels in input image - Adversarial methods and Activation visualization, Convolution filter visualization - Feature-based methods. We have also shown the results produced by different methods and a comparison between CAM, GradCAM, and Guided Backpropagation.
Leibniz Universität Hannover（汉诺威莱布尼茨大学）；dSPACE GmbH（dSPACE有限公司）
In recent years, an abundance of feature attribution methods for explaining neural networks have been developed. Especially in the field of computer vision, many methods for generating saliency maps providing pixel attributions exist. However, their explanations often contradict each other and it is not clear which explanation to trust. A natural solution to this problem is the aggregation of multiple explanations. We present and compare different pixel-based aggregation schemes with the goal of generating a new explanation, whose fidelity to the model's decision is higher than each individual…
Much effort has been devoted to understanding the decisions of deep neural networks in recent years. A number of model-aware saliency methods were proposed to explain individual classification decisions by creating saliency maps. However, they are not applicable when the parameters and the gradients of the underlying models are unavailable. Recently, model-agnostic methods have also received attention. As one of them, \textit{Prediction Difference Analysis} (PDA), a probabilistic sound methodology, was proposed. In this work, we first show that PDA can suffer from saturated classifiers. The sa…
Simon Fraser University（西蒙弗雷泽大学）；The University of British Columbia（不列颠哥伦比亚大学）
Being able to explain the prediction to clinical end-users is a necessity to leverage the power of AI models for clinical decision support. For medical images, saliency maps are the most common form of explanation. The maps highlight important features for AI model's prediction. Although many saliency map methods have been proposed, it is unknown how well they perform on explaining decisions on multi-modal medical images, where each modality/channel carries distinct clinical meanings of the same underlying biomedical phenomenon. Understanding such modality-dependent features is essential for c…
Boston University（波士顿大学）；University of Illinois at Urbana-Champaign（伊利诺伊大学厄巴纳-香槟分校）；MIT-IBM Watson AI Lab（麻省理工学院-IBM沃森人工智能实验室）
Explaining a deep learning model can help users understand its behavior and allow researchers to discern its shortcomings. Recent work has primarily focused on explaining models for tasks like image classification or visual question answering. In this paper, we introduce Salient Attributes for Network Explanation (SANE) to explain image similarity models, where a model's output is a score measuring the similarity of two inputs rather than a classification score. In this task, an explanation depends on both of the input images, so standard methods do not apply. Our SANE explanations pairs a sal…
University of Liechtenstein（列支敦士登大学）；University of Lausanne（洛桑大学）
Interpreting a large number of neurons in deep learning is difficult. Our proposed `CLAssifier-DECoder' architecture (ClaDec) facilitates the understanding of the output of an arbitrary layer of neurons or subsets thereof. It uses a decoder that transforms the incomprehensible representation of the given neurons to a representation that is more similar to the domain a human is familiar with. In an image recognition problem, one can recognize what information (or concepts) a layer maintains by contrasting reconstructed images of ClaDec with those of a conventional auto-encoder(AE) serving as re…
Deep convolutional networks have been quite successful at various image classification tasks. The current methods to explain the predictions of a pre-trained model rely on gradient information, often resulting in saliency maps that focus on the foreground object as a whole. However, humans typically reason by dissecting an image and pointing out the presence of smaller concepts. The final output is often an aggregation of the presence or absence of these smaller concepts. In this work, we propose MACE: a Model Agnostic Concept Extractor, which can explain the working of a convolutional network…
Explaining a classification result produced by an image- and video-classification model is one of the important but challenging issues in computer vision. Many methods have been proposed for producing heat-map--based explanations for this purpose, including ones based on the white-box approach that uses the internal information of a model (e.g., LRP, Grad-CAM, and Grad-CAM++) and ones based on the black-box approach that does not use any internal information (e.g., LIME, SHAP, and RISE). We propose a new black-box method BOREx (Bayesian Optimization for Refinement of visual model Explanation) …
Revealing the Gap in Human and VLM Scene Description through Counterfactual Semantic Saliency
Department of Computer Science, University of California, Santa Barbara（加州大学圣巴巴拉分校计算机科学系）；Department of Psychological and Brain Sciences, University of California, Santa Barbara（加州大学圣巴巴拉分校心理学与脑科学系）
University of Pittsburgh（匹兹堡大学）；Carnegie Mellon University（卡内基梅隆大学）；University of Pittsburgh Medical School（匹兹堡大学医学院）
We propose a BlackBox Counterfactual Explainer, designed to explain image classification models for medical applications. Classical approaches (e.g., saliency maps) that assess feature importance do not explain "how" imaging features in important anatomical regions are relevant to the classification decision. Our framework explains the decision for a target class by gradually "exaggerating" the semantic effect of the class in a query image. We adopted a Generative Adversarial Network (GAN) to generate a progressive set of perturbations to a query image, such that the classification decision ch…
Saliency methods generating visual explanatory maps representing the importance of image pixels for model classification is a popular technique for explaining neural network decisions. Hierarchical dynamic masks (HDM), a novel explanatory maps generation method, is proposed in this paper to enhance the granularity and comprehensiveness of saliency maps. First, we suggest the dynamic masks (DM), which enables multiple small-sized benchmark mask vectors to roughly learn the critical information in the image through an optimization method. Then the benchmark mask vectors guide the learning of lar…
UvA-Bosch Delta Lab（阿姆斯特丹大学-博世德尔塔实验室）；University of Amsterdam（阿姆斯特丹大学）
In this paper, we aim to explain the decisions of neural networks by utilizing multimodal information. That is counter-intuitive attributes and counter visual examples which appear when perturbed samples are introduced. Different from previous work on interpreting decisions using saliency maps, text, or visual patches we propose to use attributes and counter-attributes, and examples and counter-examples as part of the visual explanations. When humans explain visual decisions they tend to do so by providing attributes and examples. Hence, inspired by the way of human explanations in this paper …
University of Munich（慕尼黑大学）；Siemens AG, Corporate Technology（西门子股份公司，企业技术部）
Capsule Networks, as alternatives to Convolutional Neural Networks, have been proposed to recognize objects from images. The current literature demonstrates many advantages of CapsNets over CNNs. However, how to create explanations for individual classifications of CapsNets has not been well explored. The widely used saliency methods are mainly proposed for explaining CNN-based classifications; they create saliency map explanations by combining activation values and the corresponding gradients, e.g., Grad-CAM. These saliency methods require a specific architecture of the underlying classifiers…
Tsinghua University（清华大学）；Beijing National Research Center for Information Science and Technology（北京信息科学与技术国家研究中心）
Explaining deep convolutional neural networks has been recently drawing increasing attention since it helps to understand the networks' internal operations and why they make certain decisions. Saliency maps, which emphasize salient regions largely connected to the network's decision-making, are one of the most common ways for visualizing and analyzing deep networks in the computer vision community. However, saliency maps generated by existing methods cannot represent authentic information in images due to the unproven proposals about the weights of activation maps which lack solid theoretical …
