对抗训练已被证明是提升深度神经网络抵御对抗攻击鲁棒性的一种高度有效方法。然而，人们观察到它在鲁棒公平性方面存在局限，表现为不同类别之间的鲁棒性存在显著差异。近期缓解该问题的努力转向了类别重加权方法。但是，这些方法缺乏严格的理论分析，并且对权重空间的探索有限，因为它们主要依赖现有启发式算法或直觉来计算权重。此外，由于权重与模型参数的解耦优化，这些方法无法保证优化方向的一致性。它们可能导致次优的权重分配，进而产生次优模型。为解决这些问题，本文提出一种新颖的 min-max 训练框架 Class Optimal Distribution Adversarial Training（CODAT，类别最优分布对抗训练），该框架采用分布鲁棒优化来充分探索类别权重空间，从而能够在具有理论保证的情况下识别最优权重。此外，我们推导了内部最大化问题的闭式最优解，进而得到一个确定性等价目标函数，为权重与模型参数的联合优化提供理论基础。同时，我们提出公平弹性系数，用于从鲁棒性和鲁棒公平性两方面评估算法。在多个数据集上的实验结果表明，所提方法能够有效提升模型的鲁棒公平性，并优于最先进方法。
英文摘要：
Adversarial training has proven to be a highly effective method for improving the robustness of deep neural networks against adversarial attacks. Nonetheless, it has been observed to exhibit a limitation in terms of robust fairness, characterized by a significant disparity in robustness across different classes. Recent efforts to mitigate this problem have turned to class-wise reweighted methods. However, these methods suffer from a lack of rigorous theoretical analysis and are limited in their exploration of the weight space, as they mainly rely on existing heuristic algorithms or intuition to compute weights. In addition, these methods fail to guarantee the consistency of the optimization direction due to the decoupled optimization of weights and the model parameters. They potentially lead to suboptimal weight assignments and consequently, a suboptimal model. To address these problems, this paper proposes a novel min-max training framework, Class Optimal Distribution Adversarial Training (CODAT), which employs distributionally robust optimization to fully explore the class-wise weight space, thus enabling the identification of the optimal weight with theoretical guarantees. Furthermore, we derive a closed-form optimal solution to the internal maximization and then get a deterministic equivalent objective function, which provides a theoretical basis for the joint optimization of weights and model parameters. Meanwhile, we propose a fairness elasticity coefficient for the evaluation of the algorithm with regard to both robustness and robust fairness. Experimental results on various datasets show that the proposed method can effectively improve the robust fairness of the model and outperform the state-of-the-art approaches.
Korea Advanced Institute of Science and Technology (KAIST)（韩国科学技术院）
Despite their overwhelming success on a wide range of applications, convolutional neural networks (CNNs) are widely recognized to be vulnerable to adversarial examples. This intriguing phenomenon led to a competition between adversarial attacks and defense techniques. So far, adversarial training is the most widely used method for defending against adversarial attacks. It has also been extended to defend against universal adversarial perturbations (UAPs). The SOTA universal adversarial training (UAT) method optimizes a single perturbation for all training samples in the mini-batch. In this wor…
Adversarial training has been widely acknowledged as the most effective method to improve the adversarial robustness against adversarial examples for Deep Neural Networks (DNNs). So far, most existing works focus on enhancing the overall model robustness, treating each class equally in both the training and testing phases. Although revealing the disparity in robustness among classes, few works try to make adversarial training fair at the class level without sacrificing overall robustness. In this paper, we are the first to theoretically and empirically investigate the preference of different c…
Machine learning fairness concerns about the biases towards certain protected or sensitive group of people when addressing the target tasks. This paper studies the debiasing problem in the context of image classification tasks. Our data analysis on facial attribute recognition demonstrates (1) the attribution of model bias from imbalanced training data distribution and (2) the potential of adversarial examples in balancing data distribution. We are thus motivated to employ adversarial example to augment the training data for visual debiasing. Specifically, to ensure the adversarial generalizat…
Khoury College of Computer Sciences（胡里计算机学院）；Northeastern University（东北大学）
Motivated by the need for fair algorithmic decision making in the age of automation and artificially-intelligent technology, this technical report provides a theoretical insight into adversarial training for fairness in deep learning. We build upon previous work in adversarial fairness, show the persistent tradeoff between fair predictions and model performance, and explore further mechanisms that help in offsetting this tradeoff.
CuraCloud Corporation（库拉云公司）；University at Albany, State University of New York（纽约州立大学奥尔巴尼分校）；Chengdu University of Information Technology（成都信息工程大学）
Deep neural networks have been demonstrated to be vulnerable to adversarial attacks: subtle perturbations can completely change the classification results. Their vulnerability has led to a surge of research in this direction. However, most works dedicated to attacking anchor-based object detection models. In this work, we aim to present an effective and efficient algorithm to generate adversarial examples to attack anchor-free object models based on two approaches. First, we conduct category-wise instead of instance-wise attacks on the object detectors. Second, we leverage the high-level seman…
In this paper, we take a first step towards answering the question of how to design fair machine learning algorithms that are robust to adversarial attacks. Using a minimax framework, we aim to design an adversarially robust fair regression model that achieves optimal performance in the presence of an attacker who is able to add a carefully designed adversarial data point to the dataset or perform a rank-one attack on the dataset. By solving the proposed nonsmooth nonconvex-nonconcave minimax problem, the optimal adversary as well as the robust fairness-aware regression model are obtained. For…
Adversarial training algorithms have been proved to be reliable to improve machine learning models' robustness against adversarial examples. However, we find that adversarial training algorithms tend to introduce severe disparity of accuracy and robustness between different groups of data. For instance, a PGD adversarially trained ResNet18 model on CIFAR-10 has 93% clean accuracy and 67% PGD l-infty-8 robust accuracy on the class "automobile" but only 65% and 17% on the class "cat". This phenomenon happens in balanced datasets and does not exist in naturally trained models when only using clea…
SenseTime Research（商汤科技研究院）；Institute of Information Engineering, Chinese Academy of Sciences（中国科学院信息工程研究所）；Beihang University（北京航空航天大学）
Adversarial training (AT) methods are effective against adversarial attacks, yet they introduce severe disparity of accuracy and robustness between different classes, known as the robust fairness problem. Previously proposed Fair Robust Learning (FRL) adaptively reweights different classes to improve fairness. However, the performance of the better-performed classes decreases, leading to a strong performance drop. In this paper, we observed two unfair phenomena during adversarial training: different difficulties in generating adversarial examples from each class (source-class fairness) and dis…
National Taiwan University（国立台湾大学）；Mobile Drive Technology（移动驱动科技公司）
Fairness and robustness play vital roles in trustworthy machine learning. Observing safety-critical needs in various annotation-expensive vision applications, we introduce a novel learning framework, Fair Robust Active Learning (FRAL), generalizing conventional active learning to fair and adversarial robust scenarios. This framework allows us to achieve standard and robust minimax fairness with limited acquired labels. In FRAL, we then observe existing fairness-aware data selection strategies suffer from either ineffectiveness under severe data imbalance or inefficiency due to huge computation…
Dalian University of Technology（大连理工大学）；The University of Iowa（爱荷华大学）；Netflix（奈飞公司）；Texas A&M University（得克萨斯农工大学）
Improving fairness between privileged and less-privileged sensitive attribute groups (e.g, {race, gender}) has attracted lots of attention. To enhance the model performs uniformly well in different sensitive attributes, we propose a principled \underline{R}obust \underline{A}dversarial \underline{A}ttribute \underline{N}eighbourhood (RAAN) loss to debias the classification head and promote a fairer representation distribution across different sensitive attribute groups. The key idea of RAAN is to mitigate the differences of biased representations between different sensitive attribute groups by…
Duke University（杜克大学）；Monta Vista High School（蒙塔维斯塔高中）
Neural networks are getting better accuracy with higher energy and computational cost. After quantization, the cost can be greatly saved, and the quantized models are more hardware friendly with acceptable accuracy loss. On the other hand, recent research has found that neural networks are vulnerable to adversarial attacks, and the robustness of a neural network model can only be improved with defense methods, such as adversarial training. In this work, we find that adversarially-trained neural networks are more vulnerable to quantization loss than plain models. To minimize both the adversaria…
Korea Advanced Institute of Science and Technology (KAIST)（韩国科学技术院）
Convolutional neural networks (CNNs) have made significant advancement, however, they are widely known to be vulnerable to adversarial attacks. Adversarial training is the most widely used technique for improving adversarial robustness to strong white-box attacks. Prior works have been evaluating and improving the model average robustness without class-wise evaluation. The average evaluation alone might provide a false sense of robustness. For example, the attacker can focus on attacking the vulnerable class, which can be dangerous, especially, when the vulnerable class is a critical one, such…
Deep neural networks can be easily fooled into making incorrect predictions through corruption of the input by adversarial perturbations: human-imperceptible artificial noise. So far adversarial training has been the most successful defense against such adversarial attacks. This work focuses on improving adversarial training to boost adversarial robustness. We first analyze, from an instance-wise perspective, how adversarial vulnerability evolves during adversarial training. We find that during training an overall reduction of adversarial loss is achieved by sacrificing a considerable proporti…
Deep neural networks (DNNs) are shown to be susceptible to adversarial example attacks. Most existing works achieve this malicious objective by crafting subtle pixel-wise perturbations, and they are difficult to launch in the physical world due to inevitable transformations (e.g., different photographic distances and angles). Recently, there are a few research works on generating physical adversarial examples, but they generally require the details of the model a priori, which is often impractical. In this work, we propose a novel physical adversarial attack for arbitrary black-box DNN models,…
University of Belgrade - Faculty of Organizational Sciences（贝尔格莱德大学组织科学学院）；University of Belgrade - Faculty of Mathematics（贝尔格莱德大学数学学院）
With growing awareness of societal impact of artificial intelligence, fairness has become an important aspect of machine learning algorithms. The issue is that human biases towards certain groups of population, defined by sensitive features like race and gender, are introduced to the training data through data collection and labeling. Two important directions of fairness ensuring research have focused on (i) instance weighting in order to decrease the impact of more biased instances and (ii) adversarial training in order to construct data representations informative of the target variable, but…
MPI-SWS（马克斯·普朗克软件系统研究所）；University of California, Davis（加利福尼亚大学戴维斯分校）
Clustering algorithms are ubiquitous in modern data science pipelines, and are utilized in numerous fields ranging from biology to facility location. Due to their widespread use, especially in societal resource allocation problems, recent research has aimed at making clustering algorithms fair, with great success. Furthermore, it has also been shown that clustering algorithms, much like other machine learning algorithms, are susceptible to adversarial attacks where a malicious entity seeks to subvert the performance of the learning algorithm. However, despite these known vulnerabilities, there…
Tsinghua Berkeley Shenzhen Institute, Tsinghua University（清华大学清华伯克利深圳学院）；Tsinghua Shenzhen International Graduate School, Tsinghua University（清华大学深圳国际研究生院）；Key Lab. of Machine Perception (MoE), School of EECS, Peking University（北京大学信息科学技术学院机器感知与智能教育部重点实验室）；School of Information Technology, Deakin University（迪肯大学信息技术学院）；PCL Research Center of Networks and Communications, Peng Cheng Laboratory（鹏城实验室网络通信研究中心）
The study of adversarial examples and their activation has attracted significant attention for secure and robust learning with deep neural networks (DNNs). Different from existing works, in this paper, we highlight two new characteristics of adversarial examples from the channel-wise activation perspective: 1) the activation magnitudes of adversarial examples are higher than that of natural examples; and 2) the channels are activated more uniformly by adversarial examples than natural examples. We find that the state-of-the-art defense adversarial training has addressed the first issue of high…
Visual Computing University of Siegen（锡根大学视觉计算学院）；Fraunhofer ITWM Kaiserslautern（弗劳恩霍夫 ITWM 凯泽斯劳滕研究所）；IMLA University of Offenburg（奥芬堡大学 IMLA 学院）；Max Planck Institute for Informatics（马克斯·普朗克信息学研究所）
本文分析对抗训练鲁棒分类器的类别间偏差，提出 Class False Positive Score，揭示目标类别误报数对攻击脆弱性的关键影响。
Towards Class-wise Fair Adversarial Training via Anti-Bias Soft Label Distillation
Institute of Artificial Intelligence, Beihang University（北京航空航天大学人工智能研究院）；School of Software, Beihang University（北京航空航天大学软件学院）；Security Department, Alibaba Group（阿里集团安全部）
We turn the definition of individual fairness on its head---rather than ascertaining the fairness of a model given a predetermined metric, we find a metric for a given model that satisfies individual fairness. This can facilitate the discussion on the fairness of a model, addressing the issue that it may be difficult to specify a priori a suitable metric. Our contributions are twofold: First, we introduce the definition of a minimal metric and characterize the behavior of models in terms of minimal metrics. Second, for more complicated models, we apply the mechanism of randomized smoothing fro…
Causal Fair Metric: Bridging Causality, Individual Fairness, and Adversarial Robustness
Max Planck Institute for Intelligent Systems（马克斯·普朗克智能系统研究所）；Tübingen AI Center（蒂宾根人工智能中心）；McGill University（麦吉尔大学）；University of Montréal（蒙特利尔大学）；MILA（米拉计算科学研究所）
École Polytechnique Fédérale de Lausanne (EPFL)（洛桑联邦理工学院）
State-of-the-art methods for counting people in crowded scenes rely on deep networks to estimate crowd density. While effective, deep learning approaches are vulnerable to adversarial attacks, which, in a crowd-counting context, can lead to serious security issues. However, attack and defense mechanisms have been virtually unexplored in regression tasks, let alone for crowd density estimation. In this paper, we investigate the effectiveness of existing attack strategies on crowd-counting networks, and introduce a simple yet effective pixel-wise detection mechanism. It builds on the intuition t…
Deep neural network models are used today in various applications of artificial intelligence, the strengthening of which, in the face of adversarial attacks is of particular importance. An appropriate solution to adversarial attacks is adversarial training, which reaches a trade-off between robustness and generalization. This paper introduces a novel framework (Layer Sustainability Analysis (LSA)) for the analysis of layer vulnerability in an arbitrary neural network in the scenario of adversarial attacks. LSA can be a helpful toolkit to assess deep neural networks and to extend the adversaria…
IBM Research – T. J. Watson Research Center（IBM研究院–托马斯·J·沃森研究中心）；DataArt（数据艺术公司）；IBM Research – Africa（IBM研究院–非洲分院）
To ensure trust in AI models, it is becoming increasingly apparent that evaluation of models must be extended beyond traditional performance metrics, like accuracy, to other dimensions, such as fairness, explainability, adversarial robustness, and distribution shift. We describe an empirical study to evaluate multiple model types on various metrics along these dimensions on several datasets. Our results show that no particular model type performs well on all dimensions, and demonstrate the kinds of trade-offs involved in selecting models evaluated along multiple dimensions.
University of Texas at El Paso（德克萨斯大学埃尔帕索分校）；Virginia Tech（弗吉尼亚理工大学）；Austin Peay State University（奥斯汀佩伊州立大学）
In this research, we focus on the usage of adversarial sampling to test for the fairness in the prediction of deep neural network model across different classes of image in a given dataset. While several framework had been proposed to ensure robustness of machine learning model against adversarial attack, some of which includes adversarial training algorithm. There is still the pitfall that adversarial training algorithm tends to cause disparity in accuracy and robustness among different group. Our research is aimed at using adversarial sampling to test for fairness in the prediction of deep n…
The University of Electronic Science and Technology of China（电子科技大学）
Although great progress has been made on adversarial attacks for deep neural networks (DNNs), their transferability is still unsatisfactory, especially for targeted attacks. There are two problems behind that have been long overlooked: 1) the conventional setting of $T$ iterations with the step size of $ε/T$ to comply with the $ε$-constraint. In this case, most of the pixels are allowed to add very small noise, much less than $ε$; and 2) usually manipulating pixel-wise noise. However, features of a pixel extracted by DNNs are influenced by its surrounding regions, and different DNNs generally …
University of Tübingen（蒂宾根大学）；Max Planck Institute for Intelligent Systems（马克斯·普朗克智能系统研究所）
Algorithmic recourse aims to disclose the inner workings of the black-box decision process in situations where decisions have significant consequences, by providing recommendations to empower beneficiaries to achieve a more favorable outcome. To ensure an effective remedy, suggested interventions must not only be low-cost but also robust and fair. This goal is accomplished by providing similar explanations to individuals who are alike. This study explores the concept of individual fairness and adversarial robustness in causal algorithmic recourse and addresses the challenge of achieving both. …
How can we learn a classifier that is "fair" for a protected or sensitive group, when we do not know if the input to the classifier belongs to the protected group? How can we train such a classifier when data on the protected group is difficult to attain? In many settings, finding out the sensitive input attribute can be prohibitively expensive even during model training, and sometimes impossible during model serving. For example, in recommender systems, if we want to predict if a user will click on a given recommendation, we often do not know many attributes of the user, e.g., race or age, an…
University of Trieste（的里雅斯特大学）；Saarland University（萨尔大学）；University of Edinburgh（爱丁堡大学）；SISSA（国际高等研究学院）
We consider the problem of the stability of saliency-based explanations of Neural Network predictions under adversarial attacks in a classification task. Saliency interpretations of deterministic Neural Networks are remarkably brittle even when the attacks fail, i.e. for attacks that do not change the classification label. We empirically show that interpretations provided by Bayesian Neural Networks are considerably more stable under adversarial perturbations of the inputs and even under direct attacks to the explanations. By leveraging recent results, we also provide a theoretical explanation…
On the Fairness ROAD: Robust Optimization for Adversarial Debiasing
AXA Group Operations（安盛集团运营部）；Stanford University（斯坦福大学）；TRAIL, Sorbonne Université（索邦大学TRAIL）；LERIA, Université d’Angers（昂热大学LERIA）；Polish Academy of Science（波兰科学院）
Automatic Feature Fairness in Recommendation via Adversaries
National University of Singapore（新加坡国立大学）；Nanyang Technological University（南洋理工大学）；UC, San Diego（加州大学圣地亚哥分校）；AWS AI Research & Education（AWS人工智能研究与教育部门）
Recent works reveal that re-calibrating the intermediate activation of adversarial examples can improve the adversarial robustness of a CNN model. The state of the arts [Baiet al., 2021] and [Yanet al., 2021] explores this feature at the channel level, i.e. the activation of a channel is uniformly scaled by a factor. In this paper, we investigate the intermediate activation manipulation at a more fine-grained level. Instead of uniformly scaling the activation, we individually adjust each element within an activation and thus propose Element-Wise Activation Scaling, dubbed EWAS, to improve CNNs…
Almost all current adversarial attacks of CNN classifiers rely on information derived from the output layer of the network. This work presents a new adversarial attack based on the modeling and exploitation of class-wise and layer-wise deep feature distributions. We achieve state-of-the-art targeted blackbox transfer-based attack results for undefended ImageNet models. Further, we place a priority on explainability and interpretability of the attacking process. Our methodology affords an analysis of how adversarial attacks change the intermediate feature distributions of CNNs, as well as a mea…
Adversarial Pruning: A Survey and Benchmark of Pruning Methods for Adversarial Robustness
University of Cagliari（卡利亚里大学）；Sapienza University of Rome（罗马大学）；University of Genova（热那亚大学）；CINI, National Cybersecurity Lab, Consorzio Interuniversitario Nazionale per l’Informatica（意大利国家跨大学信息学联盟国家网络安全实验室）
Intuit Tech Futures（财捷科技未来部门）；Bar Ilan University（巴伊兰大学）；Intuit（财捷公司）
Latent factor models for recommender systems represent users and items as low dimensional vectors. Privacy risks of such systems have previously been studied mostly in the context of recovery of personal information in the form of usage records from the training data. However, the user representations themselves may be used together with external data to recover private user information such as gender and age. In this paper we show that user vectors calculated by a common recommender system can be exploited in this way. We propose the privacy-adversarial framework to eliminate such leakage of …
East China Normal University（华东师范大学）；Nanyang Technological University（南洋理工大学）；Alibaba Group（阿里巴巴集团）；University of Alberta（阿尔伯塔大学）；Zhejiang Sci-Tech University（浙江理工大学）
High-level representation-guided pixel denoising and adversarial training are independent solutions to enhance the robustness of CNNs against adversarial attacks by pre-processing input data and re-training models, respectively. Most recently, adversarial training techniques have been widely studied and improved while the pixel denoising-based method is getting less attractive. However, it is still questionable whether there exists a more advanced pixel denoising-based method and whether the combination of the two solutions benefits each other. To this end, we first comprehensively investigate…
College of Intelligence and Computing（智能与计算学院）；Tianjin University（天津大学）；School of Cyberspace Science and Technology（网络空间安全学院）；Beijing Institute of Technology（北京理工大学）；National Key Laboratory of Science and Technology on Information System Security（信息系统安全科学与技术国家重点实验室）
Vision transformers (ViTs) have demonstrated impressive performance and stronger adversarial robustness compared to Convolutional Neural Networks (CNNs). On the one hand, ViTs' focus on global interaction between individual patches reduces the local noise sensitivity of images. On the other hand, the neglect of noise sensitivity differences between image regions by existing decision-based attacks further compromises the efficiency of noise compression, especially for ViTs. Therefore, validating the black-box adversarial robustness of ViTs when the target model can only be queried still remains…
