现有对抗攻击往往忽视扰动稀疏性，限制了其对结构性变化进行建模以及解释深度神经网络（DNN）如何处理有意义输入模式的能力。我们提出ATOS（Attack Through Overlapping Sparsity，通过重叠稀疏性进行攻击），这是一个可微优化框架，能够以元素级、像素级和分组级形式生成结构化、稀疏的对抗扰动。针对图像分类器的白盒攻击，我们引入Overlapping Smoothed L0（OSL0，重叠平滑L0）函数，该函数促进收敛到驻点，同时鼓励稀疏、结构化的扰动。通过对通道和相邻像素进行分组，ATOS提高了可解释性，并有助于识别鲁棒特征与非鲁棒特征。我们使用指数绝对值之和的对数来近似L-infinity梯度，以严格控制扰动幅度。在CIFAR-10和ImageNet上，ATOS实现了100%的攻击成功率，同时与先前方法相比产生了显著更稀疏且结构更连贯的扰动。结构化分组攻击从网络的视角突出关键区域，通过用来自目标类别的鲁棒特征替换定义类别的区域来提供反事实解释。
英文摘要：
Existing adversarial attacks often neglect perturbation sparsity, limiting their ability to model structural changes and to explain how deep neural networks (DNNs) process meaningful input patterns. We propose ATOS (Attack Through Overlapping Sparsity), a differentiable optimization framework that generates structured, sparse adversarial perturbations in element-wise, pixel-wise, and group-wise forms. For white-box attacks on image classifiers, we introduce the Overlapping Smoothed L0 (OSL0) function, which promotes convergence to a stationary point while encouraging sparse, structured perturbations. By grouping channels and adjacent pixels, ATOS improves interpretability and helps identify robust versus non-robust features. We approximate the L-infinity gradient using the logarithm of the sum of exponential absolute values to tightly control perturbation magnitude. On CIFAR-10 and ImageNet, ATOS achieves a 100% attack success rate while producing significantly sparser and more structurally coherent perturbations than prior methods. The structured group-wise attack highlights critical regions from the network's perspective, providing counterfactual explanations by replacing class-defining regions with robust features from the target class.
We propose a versatile framework based on random search, Sparse-RS, for score-based sparse targeted and untargeted attacks in the black-box setting. Sparse-RS does not rely on substitute models and achieves state-of-the-art success rate and query efficiency for multiple sparse attack models: $l_0$-bounded perturbations, adversarial patches, and adversarial frames. The $l_0$-version of untargeted Sparse-RS outperforms all black-box and even all white-box attacks for different models on MNIST, CIFAR-10, and ImageNet. Moreover, our untargeted Sparse-RS achieves very high success rates even for th…
We provide a methodology, resilient feature engineering, for creating adversarially resilient classifiers. According to existing work, adversarial attacks identify weakly correlated or non-predictive features learned by the classifier during training and design the adversarial noise to utilize these features. Therefore, highly predictive features should be used first during classification in order to determine the set of possible output labels. Our methodology focuses the problem of designing resilient classifiers into a problem of designing resilient feature extractors for these highly predic…
Wuhan University（武汉大学）；Pennsylvania State University（宾夕法尼亚州立大学）
Adversarial examples are well-designed input samples, in which perturbations are imperceptible to the human eyes, but easily mislead the output of deep neural networks (DNNs). Existing works synthesize adversarial examples by leveraging simple metrics to penalize perturbations, that lack sufficient consideration of the human visual system (HVS), which produces noticeable artifacts. To explore why the perturbations are visible, this paper summarizes four primary factors affecting the perceptibility of human eyes. Based on this investigation, we design a multi-factor metric MulFactorLoss for mea…
Text-guided image generation models can be prompted to generate images using nonce words adversarially designed to robustly evoke specific visual concepts. Two approaches for such generation are introduced: macaronic prompting, which involves designing cryptic hybrid words by concatenating subword units from different languages; and evocative prompting, which involves designing nonce words whose broad morphological features are similar enough to that of existing words to trigger robust visual associations. The two methods can also be combined to generate images associated with more specific vi…
Boston University（波士顿大学）；Northeastern University（东北大学）；IBM Research（IBM研究院）
Designing effective defense against adversarial attacks is a crucial topic as deep neural networks have been proliferated rapidly in many security-critical domains such as malware detection and self-driving cars. Conventional defense methods, although shown to be promising, are largely limited by their single-source single-cost nature: The robustness promotion tends to plateau when the defenses are made increasingly stronger while the cost tends to amplify. In this paper, we study principles of designing multi-source and multi-cost schemes where defense performance is boosted from multiple def…
State-of-art deep neural networks (DNN) are vulnerable to attacks by adversarial examples: a carefully designed small perturbation to the input, that is imperceptible to human, can mislead DNN. To understand the root cause of adversarial examples, we quantify the probability of adversarial example existence for linear classifiers. Previous mathematical definition of adversarial examples only involves the overall perturbation amount, and we propose a more practical relevant definition of strong adversarial examples that separately limits the perturbation along the signal direction also. We show…
This paper investigates recently proposed approaches for defending against adversarial examples and evaluating adversarial robustness. We motivate 'adversarial risk' as an objective for achieving models robust to worst-case inputs. We then frame commonly used attacks and evaluation metrics as defining a tractable surrogate objective to the true adversarial risk. This suggests that models may optimize this surrogate rather than the true adversarial risk. We formalize this notion as 'obscurity to an adversary,' and develop tools and heuristics for identifying obscured models and designing transp…
In spite of the enormous success of neural networks, adversarial examples remain a relatively weakly understood feature of deep learning systems. There is a considerable effort in both building more powerful adversarial attacks and designing methods to counter the effects of adversarial examples. We propose a method to transform the adversarial input data through a mixture of generators in order to recover the correct class obfuscated by the adversarial attack. A canonical set of images is used to generate adversarial examples through potentially multiple attacks. Such transformed images are p…
Explainable machine learning has become increasingly prevalent, especially in healthcare where explainable models are vital for ethical and trusted automated decision making. Work on the susceptibility of deep learning models to adversarial attacks has shown the ease of designing samples to mislead a model into making incorrect predictions. In this work, we propose a model agnostic explainability-based method for the accurate detection of adversarial samples on two datasets with different complexity and properties: Electronic Health Record (EHR) and chest X-ray (CXR) data. On the MIMIC-III and…
Waseda University（早稻田大学）；The University of Electro-Communications（电气通信大学）；University of Tsukuba（筑波大学）；RIKEN AIP（理化学研究所革新智能统合研究中心）
This work aims to assess the reality and feasibility of the adversarial attack against cardiac diagnosis system powered by machine learning algorithms. To this end, we introduce adversarial beats, which are adversarial perturbations tailored specifically against electrocardiograms (ECGs) beat-by-beat classification system. We first formulate an algorithm to generate adversarial examples for the ECG classification neural network model, and study its attack success rate. Next, to evaluate its feasibility in a physical environment, we mount a hardware attack by designing a malicious signal genera…
Fuzhou University（福州大学）；City University of Macau（澳门城市大学）；Peking University（北京大学）
Transferability of adversarial examples is of critical importance to launch black-box adversarial attacks, where attackers are only allowed to access the output of the target model. However, under such a challenging but practical setting, the crafted adversarial examples are always prone to overfitting to the proxy model employed, presenting poor transferability. In this paper, we suggest alleviating the overfitting issue from a novel perspective, i.e., designing a fitted model architecture. Specifically, delving the bottom of the cause of poor transferability, we arguably decompose and recons…
Deep models, while being extremely versatile and accurate, are vulnerable to adversarial attacks: slight perturbations that are imperceptible to humans can completely flip the prediction of deep models. Many attack and defense mechanisms have been proposed, although a satisfying solution still largely remains elusive. In this work, we give strong evidence that during training, deep models maximize the minimum margin in order to achieve high accuracy, but at the same time decrease the \emph{average} margin hence hurting robustness. Our empirical results highlight an intrinsic trade-off between …
Medical imaging AI systems such as disease classification and segmentation are increasingly inspired and transformed from computer vision based AI systems. Although an array of adversarial training and/or loss function based defense techniques have been developed and proved to be effective in computer vision, defending against adversarial attacks on medical images remains largely an uncharted territory due to the following unique challenges: 1) label scarcity in medical images significantly limits adversarial generalizability of the AI system; 2) vastly similar and dominant fore- and backgroun…
School of Computer and Communication Sciences, EPFL（EPFL计算机与通信科学学院）
We prove an exponential separation for the sample complexity between the standard PAC-learning model and a version of the Equivalence-Query-learning model. We then show that this separation has interesting implications for adversarial robustness. We explore a vision of designing an adaptive defense that in the presence of an attacker computes a model that is provably robust. In particular, we show how to realize this vision in a simplified setting. In order to do so, we introduce a notion of a strong adversary: he is not limited by the type of perturbations he can apply but when presented with…
Northeastern University（东北大学）；IBM Research AI（IBM研究院人工智能部门）
Deep neural networks (DNNs) are known vulnerable to adversarial attacks. That is, adversarial examples, obtained by adding delicately crafted distortions onto original legal inputs, can mislead a DNN to classify them as any target labels. In a successful adversarial attack, the targeted mis-classification should be achieved with the minimal distortion added. In the literature, the added distortions are usually measured by L0, L1, L2, and L infinity norms, namely, L0, L1, L2, and L infinity attacks, respectively. However, there lacks a versatile framework for all types of adversarial attacks. T…
An important goal in deep learning is to learn versatile, high-level feature representations of input data. However, standard networks' representations seem to possess shortcomings that, as we illustrate, prevent them from fully realizing this goal. In this work, we show that robust optimization can be re-cast as a tool for enforcing priors on the features learned by deep neural networks. It turns out that representations learned by robust models address the aforementioned shortcomings and make significant progress towards learning a high-level encoding of inputs. In particular, these represen…
University of Michigan（密歇根大学）；NVIDIA（英伟达）；California Institute of Technology（加州理工学院）；ASU（亚利桑那州立大学）；Stanford University（斯坦福大学）
Trajectory prediction using deep neural networks (DNNs) is an essential component of autonomous driving (AD) systems. However, these methods are vulnerable to adversarial attacks, leading to serious consequences such as collisions. In this work, we identify two key ingredients to defend trajectory prediction models against adversarial attacks including (1) designing effective adversarial training methods and (2) adding domain-specific data augmentation to mitigate the performance degradation on clean data. We demonstrate that our method is able to improve the performance by 46% on adversarial …
Model-Free Adversarial Purification via Coarse-To-Fine Tensor Network Representation
RIKEN Center for Advanced Intelligence Project (RIKEN AIP)（理化学研究所先进智能项目中心（RIKEN AIP））；Tokyo University of Agriculture and Technology（东京农工大学）；Institute of Science Tokyo（东京科学大学）
Adversarial training for neural networks has been in the limelight in recent years. The advancement in neural network architectures over the last decade has led to significant improvement in their performance. It sparked an interest in their deployment for real-time applications. This process initiated the need to understand the vulnerability of these models to adversarial attacks. It is instrumental in designing models that are robust against adversaries. Recent works have proposed novel techniques to counter the adversaries, most often sacrificing natural accuracy. Most suggest training with…
Adversarial extraction attacks constitute an insidious threat against Deep Learning (DL) models in-which an adversary aims to steal the architecture, parameters, and hyper-parameters of a targeted DL model. Existing extraction attack literature have observed varying levels of attack success for different DL models and datasets, yet the underlying cause(s) behind their susceptibility often remain unclear, and would help facilitate creating secure DL systems. In this paper we present PINCH: an efficient and automated extraction attack framework capable of designing, deploying, and analyzing extr…
National Technical University of Athens（雅典国立技术大学）；Athena Research Center（雅典娜研究中心）
Designing powerful adversarial attacks is of paramount importance for the evaluation of $\ell_p$-bounded adversarial defenses. Projected Gradient Descent (PGD) is one of the most effective and conceptually simple algorithms to generate such adversaries. The search space of PGD is dictated by the steepest ascent directions of an objective. Despite the plethora of objective function choices, there is no universally superior option and robustness overestimation may arise from ill-suited objective selection. Driven by this observation, we postulate that the combination of different objectives thro…
Recently, there has been an abundance of works on designing Deep Neural Networks (DNNs) that are robust to adversarial examples. In particular, a central question is which features of DNNs influence adversarial robustness and, therefore, can be to used to design robust DNNs. In this work, this problem is studied through the lens of compression which is captured by the low-rank structure of weight matrices. It is first shown that adversarial training tends to promote simultaneously low-rank and sparse structure in the weight matrices of neural networks. This is measured through the notions of e…
The interplay between quantum physics and machine learning gives rise to the emergent frontier of quantum machine learning, where advanced quantum learning models may outperform their classical counterparts in solving certain challenging problems. However, quantum learning systems are vulnerable to adversarial attacks: adding tiny carefully-crafted perturbations on legitimate input samples can cause misclassifications. To address this issue, we propose a general scheme to protect quantum learning systems from adversarial attacks by randomly encoding the legitimate data samples through unitary …
USI（瑞士意大利语大学）；NNAISENSE；Imperial College London（帝国理工学院）；Intel Perceptual Computing（英特尔感知计算）；Fabula AI（Fabula人工智能）；Stanford University（斯坦福大学）
Deep learning systems have become ubiquitous in many aspects of our lives. Unfortunately, it has been shown that such systems are vulnerable to adversarial attacks, making them prone to potential unlawful uses. Designing deep neural networks that are robust to adversarial attacks is a fundamental step in making such systems safer and deployable in a broader variety of applications (e.g. autonomous driving), but more importantly is a necessary step to design novel and more advanced architectures built on new computational paradigms rather than marginally building on the existing ones. In this p…
Hangzhou Innovation Institute, Beihang University（北京航空航天大学杭州创新研究院）；Beihang University（北京航空航天大学）；University of Ottawa（渥太华大学）
Recent researches have suggested that the predictive accuracy of neural network may contend with its adversarial robustness. This presents challenges in designing effective regularization schemes that also provide strong adversarial robustness. Revisiting Vicinal Risk Minimization (VRM) as a unifying regularization principle, we propose Adversarial Labelling of Perturbed Samples (ALPS) as a regularization scheme that aims at improving the generalization ability and adversarial robustness of the trained model. ALPS trains neural networks with synthetic samples formed by perturbing each authenti…
Deep neural networks for natural language processing tasks are vulnerable to adversarial input perturbations. In this paper, we present a versatile language for programmatically specifying string transformations -- e.g., insertions, deletions, substitutions, swaps, etc. -- that are relevant to the task at hand. We then present an approach to adversarially training models that are robust to such user-defined string transformations. Our approach combines the advantages of search-based techniques for adversarial training with abstraction-based techniques. Specifically, we show how to decompose a …
SenseTime（商汤科技）；Zhongguancun Laboratory（中关村实验室）；Hefei Comprehensive National Science Center（合肥综合性国家科学中心）；JD Explore Academy（京东探索研究院）；Beihang University（北京航空航天大学）；Chinese Academy of Sciences（中国科学院）；Zhejiang University（浙江大学）
Adversarial training has been demonstrated to be one of the most effective remedies for defending adversarial examples, yet it often suffers from the huge robustness generalization gap on unseen testing adversaries, deemed as the adversarially robust generalization problem. Despite the preliminary understandings devoted to adversarially robust generalization, little is known from the architectural perspective. To bridge the gap, this paper for the first time systematically investigated the relationship between adversarially robust generalization and architectural design. Inparticular, we compr…
Tsinghua University（清华大学）；PCL Research Center of Networks and Communications, Peng Cheng Laboratory（鹏城实验室网络通信PCL研究中心）；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教育部重点实验室）
The study on improving the robustness of deep neural networks against adversarial examples grows rapidly in recent years. Among them, adversarial training is the most promising one, which flattens the input loss landscape (loss change with respect to input) via training on adversarially perturbed examples. However, how the widely used weight loss landscape (loss change with respect to weight) performs in adversarial training is rarely explored. In this paper, we investigate the weight loss landscape from a new perspective, and identify a clear correlation between the flatness of weight loss la…
While vision-language pre-training model (VLP) has shown revolutionary improvements on various vision-language (V+L) tasks, the studies regarding its adversarial robustness remain largely unexplored. This paper studied the adversarial attack on popular VLP models and V+L tasks. First, we analyzed the performance of adversarial attacks under different settings. By examining the influence of different perturbed objects and attack targets, we concluded some key observations as guidance on both designing strong multimodal adversarial attack and constructing robust VLP models. Second, we proposed a…
We consider the theoretical problem of designing an optimal adversarial attack on a decision system that maximally degrades the achievable performance of the system as measured by the mutual information between the degraded signal and the label of interest. This problem is motivated by the existence of adversarial examples for machine learning classifiers. By adopting an information theoretic perspective, we seek to identify conditions under which adversarial vulnerability is unavoidable i.e. even optimally designed classifiers will be vulnerable to small adversarial perturbations. We present …
We propose a new adversarial training approach for injecting robustness when designing controllers for upcoming cyber-physical power systems. Previous approaches relying deeply on simulations are not able to cope with the rising complexity and are too costly when used online in terms of computation budget. In comparison, our method proves to be computationally efficient online while displaying useful robustness properties. To do so we model an adversarial framework, propose the implementation of a fixed opponent policy and test it on a L2RPN (Learning to Run a Power Network) environment. This …
Northeastern University（东北大学）；MIT-IBM Watson AI Lab（MIT-IBM沃森人工智能实验室）；IBM Research（IBM研究院）；Massachusetts Institute of Technology（麻省理工学院）
It is known that deep neural networks (DNNs) are vulnerable to adversarial attacks. The so-called physical adversarial examples deceive DNN-based decisionmakers by attaching adversarial patches to real objects. However, most of the existing works on physical adversarial attacks focus on static objects such as glass frames, stop signs and images attached to cardboard. In this work, we proposed adversarial T-shirts, a robust physical adversarial example for evading person detectors even if it could undergo non-rigid deformation due to a moving person's pose changes. To the best of our knowledge,…
Risk Analysis and Design Against Adversarial Actions
Dipartimento di Ingegneria dell’Informazione – Università di Brescia（信息工程系——布雷西亚大学）；NASA Langley Research Center（NASA兰利研究中心）；Dipartimento di Elettronica, Informazione e Bioingegneria – Politecnico di Milano（电子、信息与生物工程系——米兰理工学院）
University of Maryland（马里兰大学）；Intelligent Automation, a BlueHalo Company（Intelligent Automation（BlueHalo 旗下公司））；Virginia Tech（弗吉尼亚理工大学）
This paper presents channel-aware adversarial attacks against deep learning-based wireless signal classifiers. There is a transmitter that transmits signals with different modulation types. A deep neural network is used at each receiver to classify its over-the-air received signals to modulation types. In the meantime, an adversary transmits an adversarial perturbation (subject to a power budget) to fool receivers into making errors in classifying signals that are received as superpositions of transmitted signals and adversarial perturbations. First, these evasion attacks are shown to fail whe…
Vision transformers rely on a patch token based self attention mechanism, in contrast to convolutional networks. We investigate fundamental differences between these two families of models, by designing a block sparsity based adversarial token attack. We probe and analyze transformer as well as convolutional models with token attacks of varying patch sizes. We infer that transformer models are more sensitive to token attacks than convolutional models, with ResNets outperforming Transformer models by up to $\sim30\%$ in robust accuracy for single token attacks.
Princeton University（普林斯顿大学）；University of Chicago（芝加哥大学）
Localized adversarial patches aim to induce misclassification in machine learning models by arbitrarily modifying pixels within a restricted region of an image. Such attacks can be realized in the physical world by attaching the adversarial patch to the object to be misclassified, and defending against such attacks is an unsolved/open problem. In this paper, we propose a general defense framework called PatchGuard that can achieve high provable robustness while maintaining high clean accuracy against localized adversarial patches. The cornerstone of PatchGuard involves the use of CNNs with sma…
Deep neural networks (DNNs) are found to be vulnerable to adversarial attacks, and various methods have been proposed for the defense. Among these methods, adversarial training has been drawing increasing attention because of its simplicity and effectiveness. However, the performance of the adversarial training is greatly limited by the architectures of target DNNs, which often makes the resulting DNNs with poor accuracy and unsatisfactory robustness. To address this problem, we propose DSARA to automatically search for the neural architectures that are accurate and robust after adversarial tr…
Adversarial Training (AT) is crucial for obtaining deep neural networks that are robust to adversarial attacks, yet recent works found that it could also make models more vulnerable to privacy attacks. In this work, we further reveal this unsettling property of AT by designing a novel privacy attack that is practically applicable to the privacy-sensitive Federated Learning (FL) systems. Using our method, the attacker can exploit AT models in the FL system to accurately reconstruct users' private training images even when the training batch size is large. Code is available at https://github.com…
Despite their immense popularity, deep learning-based acoustic systems are inherently vulnerable to adversarial attacks, wherein maliciously crafted audios trigger target systems to misbehave. In this paper, we present SirenAttack, a new class of attacks to generate adversarial audios. Compared with existing attacks, SirenAttack highlights with a set of significant features: (i) versatile -- it is able to deceive a range of end-to-end acoustic systems under both white-box and black-box settings; (ii) effective -- it is able to generate adversarial audios that can be recognized as specific phra…
