The need to explain the output of a deep neural network classifier is now widely recognized. While previous methods typically explain a single class in the output, we advocate explaining the whole output, which is a probability distribution over multiple classes. A whole-output explanation can help a human user gain an overall understanding of model behaviour instead of only one aspect of it. It can also provide a natural framework where one can examine the evidence used to discriminate between competing classes, and thereby obtain contrastive explanations. In this paper, we propose a contrastive whole-output explanation (CWOX) method for image classification, and evaluate it using quantitative metrics and through human subject studies. The source code of CWOX is available at https://github.com/vaynexie/CWOX.
University of Maryland, Baltimore County（马里兰大学巴尔的摩县分校）；Northrop Grumman（诺斯罗普·格鲁曼公司）；University of California, Davis（加州大学戴维斯分校）
Post-hoc explanation methods, e.g., Grad-CAM, enable humans to inspect the spatial regions responsible for a particular network decision. However, it is shown that such explanations are not always consistent with human priors, such as consistency across image transformations. Given an interpretation algorithm, e.g., Grad-CAM, we introduce a novel training method to train the model to produce more consistent explanations. Since obtaining the ground truth for a desired model interpretation is not a well-defined task, we adopt ideas from contrastive self-supervised learning, and apply them to the…
University of Toronto（多伦多大学）；LG AI Research（LG人工智能研究院）
Contrastive learning has led to substantial improvements in the quality of learned embedding representations for tasks such as image classification. However, a key drawback of existing contrastive augmentation methods is that they may lead to the modification of the image content which can yield undesired alterations of its semantics. This can affect the performance of the model on downstream tasks. Hence, in this paper, we ask whether we can augment image data in contrastive learning such that the task-relevant semantic content of an image is preserved. For this purpose, we propose to leverag…
Visual explanations are logical arguments based on visual features that justify the predictions made by neural networks. Current modes of visual explanations answer questions of the form $`Why \text{ } P?'$. These $Why$ questions operate under broad contexts thereby providing answers that are irrelevant in some cases. We propose to constrain these $Why$ questions based on some context $Q$ so that our explanations answer contrastive questions of the form $`Why \text{ } P, \text{} rather \text{ } than \text{ } Q?'$. In this paper, we formalize the structure of contrastive visual explanations for…
Modern self-supervised learning algorithms typically enforce persistency of instance representations across views. While being very effective on learning holistic image and video representations, such an objective becomes sub-optimal for learning spatio-temporally fine-grained features in videos, where scenes and instances evolve through space and time. In this paper, we present Contextualized Spatio-Temporal Contrastive Learning (ConST-CL) to effectively learn spatio-temporally fine-grained video representations via self-supervision. We first design a region-based pretext task which requires …
HPCR: Holistic Proxy-based Contrastive Replay for Online Continual Learning
Harbin Institute of Technology, Shenzhen（哈尔滨工业大学（深圳））；Centre for Frontier AI Research, Institute of High Performance Computing, A*STAR（新加坡高性能计算研究院前沿人工智能研究中心（新加坡科技研究局））
Yachiyo Engineering, Co.,Ltd. Research Institute for Infrastructure Paradigm Shift（八千代工程株式会社基础设施范式转换研究所）
For infrastructure inspections, damage representation does not constantly match the predefined classes of damage grade, resulting in detailed clusters of unseen damages or more complex clusters from overlapped space between two grades. The damage representation has fundamentally complex features; consequently, not all the damage classes can be perfectly predefined. The proposed MN-pair contrastive learning method helps to explore an embedding damage representation beyond the predefined classes by including more detailed clusters. It maximizes both the similarity of M-1 positive images close to…
Siemens AG（西门子股份公司）；The University of Munich（慕尼黑大学）
A number of backpropagation-based approaches such as DeConvNets, vanilla Gradient Visualization and Guided Backpropagation have been proposed to better understand individual decisions of deep convolutional neural networks. The saliency maps produced by them are proven to be non-discriminative. Recently, the Layer-wise Relevance Propagation (LRP) approach was proposed to explain the classification decisions of rectifier neural networks. In this work, we evaluate the discriminativeness of the generated explanations and analyze the theoretical foundation of LRP, i.e. Deep Taylor Decomposition. Th…
Contrastive Pretraining for Visual Concept Explanations of Socioeconomic Outcomes
Technical University of Munich（慕尼黑工业大学）；Munich Center for Machine Learning（慕尼黑机器学习中心）；Vrije Universiteit Amsterdam（阿姆斯特丹自由大学）；Getulio Vargas Foundation（热图利奥·瓦加斯基金会）；Inria（法国国家信息与自动化研究所）；University of Montpellier（蒙彼利埃大学）
Zhejiang University（浙江大学）；University of Science and Technology of China（中国科学技术大学）；Tsinghua University（清华大学）
Knowledge Distillation (KD) aims at transferring knowledge from a larger well-optimized teacher network to a smaller learnable student network.Existing KD methods have mainly considered two types of knowledge, namely the individual knowledge and the relational knowledge. However, these two types of knowledge are usually modeled independently while the inherent correlations between them are largely ignored. It is critical for sufficient student network learning to integrate both individual knowledge and relational knowledge while reserving their inherent correlation. In this paper, we propose t…
The goal of a classification model is to assign the correct labels to data. In most cases, this data is not fully described by the given set of labels. Often a rich set of meaningful concepts exist in the domain that can much more precisely describe each datapoint. Such concepts can also be highly useful for interpreting the model's classifications. In this paper we propose a model, denoted as Variational Autoencoder-based Contrastive Explanation (VAE-CE), that represents data with high-level concepts and uses this representation for both classification and generating explanations. The explana…
Stanford University（斯坦福大学）；University of California, Davis（加州大学戴维斯分校）
Training a supervised neural network classifier typically requires many annotated training samples. Collecting and annotating a large number of data points are costly and sometimes even infeasible. Traditional annotation process uses a low-bandwidth human-machine communication interface: classification labels, each of which only provides several bits of information. We propose Active Learning with Contrastive Explanations (ALICE), an expert-in-the-loop training framework that utilizes contrastive natural language explanations to improve data efficiency in learning. ALICE learns to first use ac…
IBM Research（IBM研究院）；University of Michigan（密歇根大学）
In this paper we propose a novel method that provides contrastive explanations justifying the classification of an input by a black box classifier such as a deep neural network. Given an input we find what should be %necessarily and minimally and sufficiently present (viz. important object pixels in an image) to justify its classification and analogously what should be minimally and necessarily \emph{absent} (viz. certain background pixels). We argue that such explanations are natural for humans and are used commonly in domains such as health care and criminology. What is minimally but critica…
City Data Science Institute - City, University of London（伦敦城市大学城市数据科学研究所）
A novel explainable AI method called CLEAR Image is introduced in this paper. CLEAR Image is based on the view that a satisfactory explanation should be contrastive, counterfactual and measurable. CLEAR Image explains an image's classification probability by contrasting the image with a corresponding image generated automatically via adversarial learning. This enables both salient segmentation and perturbations that faithfully determine each segment's importance. CLEAR Image was successfully applied to a medical imaging case study where it outperformed methods such as Grad-CAM and LIME by an a…
Signature Activation: A Sparse Signal View for Holistic Saliency
Harvard University（哈佛大学）；Broad Institute of Massachusetts Institute of Technology and Harvard（麻省理工学院与哈佛大学博德研究所）；Massachusetts General Hospital（麻省总医院）；Harvard Medical School（哈佛医学院）
Jadavpur University（贾达沃普尔大学）；University of Toronto（多伦多大学）
Privacy and annotation bottlenecks are two major issues that profoundly affect the practicality of machine learning-based medical image analysis. Although significant progress has been made in these areas, these issues are not yet fully resolved. In this paper, we seek to tackle these concerns head-on and systematically explore the applicability of non-contrastive self-supervised learning (SSL) algorithms under federated learning (FL) simulations for medical image analysis. We conduct thorough experimentation of recently proposed state-of-the-art non-contrastive frameworks under standard FL se…
Hint-AD: Holistically Aligned Interpretability in End-to-End Autonomous Driving
Institute for AI Industry Research (AIR)（人工智能产业研究院（AIR））；Xingjian College, Tsinghua University（清华大学行健书院）；OpenDriveLab, Shanghai AI Lab（上海人工智能实验室OpenDriveLab）；Mercedes-Benz Group China Ltd.（梅赛德斯-奔驰集团中国有限公司）
UC San Diego（加州大学圣地亚哥分校）；Stanford University（斯坦福大学）
Despite their high accuracies, modern complex image classifiers cannot be trusted for sensitive tasks due to their unknown decision-making process and potential biases. Counterfactual explanations are very effective in providing transparency for these black-box algorithms. Nevertheless, generating counterfactuals that can have a consistent impact on classifier outputs and yet expose interpretable feature changes is a very challenging task. We introduce a novel method to generate causal and yet interpretable counterfactual explanations for image classifiers using pretrained generative models wi…
University of Stuttgart（斯图加特大学）；AISIA Lab, University of Science -VNU HCM（越南国立大学胡志明市理科大学AISIA实验室）；Dongguk University（东国大学）；University of Texas at Austin（德克萨斯大学奥斯汀分校）；Max Planck Institute for Informatics（马克斯·普朗克信息学研究所）；Helmholtz AI, Helmholtz Munich（亥姆霍兹慕尼黑亥姆霍兹人工智能）；University of Bonn（波恩大学）；University of California San Diego（加利福尼亚大学圣迭戈分校）；Oldenburg University（奥尔登堡大学）；German Research Center for Artificial Intelligence（德国人工智能研究中心）
Collecting large-scale medical datasets with fully annotated samples for training of deep networks is prohibitively expensive, especially for 3D volume data. Recent breakthroughs in self-supervised learning (SSL) offer the ability to overcome the lack of labeled training samples by learning feature representations from unlabeled data. However, most current SSL techniques in the medical field have been designed for either 2D images or 3D volumes. In practice, this restricts the capability to fully leverage unlabeled data from numerous sources, which may include both 2D and 3D data. Additionally…
Beijing Institute of Technology（北京理工大学）；Inceptio Technology（嬴彻科技）
Domain adaptive semantic segmentation refers to making predictions on a certain target domain with only annotations of a specific source domain. Current state-of-the-art works suggest that performing category alignment can alleviate domain shift reasonably. However, they are mainly based on image-to-image adversarial training and little consideration is given to semantic variations of an object among images, failing to capture a comprehensive picture of different categories. This motivates us to explore a holistic representative, the semantic distribution from each category in source domain, t…
VisTA: Vision-Text Alignment Model with Contrastive Learning using Multimodal Data for Evidence-Driven, Reliable, and Explainable Alzheimer's Disease Diagnosis
Centre hospitalier universitaire vaudois (CHUV)（洛桑大学医院中心（CHUV））；Université de Lausanne (UNIL)（洛桑大学（UNIL））；VNU University of Engineering and Technology, Vietnam National University（越南国家大学下属越南工程技术大学）；University of Science, Vietnam National University Ho Chi Minh City（胡志明市越南国家大学下属科学大学）；Military Hospital 175（175军事医院）；School of Biomedical Engineering, International University, Vietnam National University（越南国家大学下属国际大学生物医学工程学院）
Institute of Automation, CAS（中国科学院自动化研究所）；School of Artificial Intelligence, UCAS（中国科学院大学人工智能学院）；Institute Of Computing Technology, CAS（中国科学院计算技术研究所）；Peng Cheng Laboratory（鹏城实验室）；Kuaishou Technology（快手科技）；National Cheng-Kung University（国立成功大学）
We present Unified Contrastive Arbitrary Style Transfer (UCAST), a novel style representation learning and transfer framework, which can fit in most existing arbitrary image style transfer models, e.g., CNN-based, ViT-based, and flow-based methods. As the key component in image style transfer tasks, a suitable style representation is essential to achieve satisfactory results. Existing approaches based on deep neural network typically use second-order statistics to generate the output. However, these hand-crafted features computed from a single image cannot leverage style information sufficient…
In this study, we use a self-explaining neural network (SENN), which learns unsupervised concepts, to acquire concepts that are easy for people to understand automatically. In concept learning, the hidden layer retains verbalizable features relevant to the output, which is crucial when adapting to real-world environments where explanations are required. However, it is known that the interpretability of concepts output by SENN is reduced in general settings, such as autonomous driving scenarios. Thus, this study combines contrastive learning with concept learning to improve the readability of c…
There has been a recent surge of research interest in attacking the problem of social relation inference based on images. Existing works classify social relations mainly by creating complicated graphs of human interactions, or learning the foreground and/or background information of persons and objects, but ignore holistic scene context. The holistic scene refers to the functionality of a place in images, such as dinning room, playground and office. In this paper, by mimicking human understanding on images, we propose an approach of \textbf{PR}actical \textbf{I}nference in \textbf{S}ocial r\te…
In this work, we introduce panoramic panoptic segmentation as the most holistic scene understanding both in terms of field of view and image level understanding for standard camera based input. A complete surrounding understanding provides a maximum of information to the agent, which is essential for any intelligent vehicle in order to make informed decisions in a safety-critical dynamic environment such as real-world traffic. In order to overcome the lack of annotated panoramic images, we propose a framework which allows model training on standard pinhole images and transfers the learned feat…
In this work, we introduce panoramic panoptic segmentation, as the most holistic scene understanding, both in terms of Field of View (FoV) and image-level understanding for standard camera-based input. A complete surrounding understanding provides a maximum of information to a mobile agent. This is essential information for any intelligent vehicle to make informed decisions in a safety-critical dynamic environment such as real-world traffic. In order to overcome the lack of annotated panoramic images, we propose a framework which allows model training on standard pinhole images and transfers t…
King Abdullah University of Science and Technology (KAUST)（阿卜杜拉国王科技大学（KAUST））
Datasets that capture the connection between vision, language, and affection are limited, causing a lack of understanding of the emotional aspect of human intelligence. As a step in this direction, the ArtEmis dataset was recently introduced as a large-scale dataset of emotional reactions to images along with language explanations of these chosen emotions. We observed a significant emotional bias towards instance-rich emotions, making trained neural speakers less accurate in describing under-represented emotions. We show that collecting new data, in the same way, is not effective in mitigating…
