Deep neural networks (DNNs) provide excellent performance across a wide range of classification tasks, but their training requires high computational resources and is often outsourced to third parties. Recent work has shown that outsourced training introduces the risk that a malicious trainer will return a backdoored DNN that behaves normally on most inputs but causes targeted misclassifications or degrades the accuracy of the network when a trigger known only to the attacker is present. In this paper, we provide the first effective defenses against backdoor attacks on DNNs. We implement three backdoor attacks from prior work and use them to investigate two promising defenses, pruning and fine-tuning. We show that neither, by itself, is sufficient to defend against sophisticated attackers. We then evaluate fine-pruning, a combination of pruning and fine-tuning, and show that it successfully weakens or even eliminates the backdoors, i.e., in some cases reducing the attack success rate to 0% with only a 0.4% drop in accuracy for clean (non-triggering) inputs. Our work provides the first step toward defenses against backdoor attacks in deep neural networks.

本文提出了针对回顾性神经网络（Badnets）的新型两级防御（NNOCULICULE），该案例在响应该字段中遇到的回溯测试输入，修复了预部署和在线的BADNET。在预部署阶段，NNICULICULE与清洁验证输入的随机扰动进行检测，以部分减少后门的对抗影响。部署后，NNOCULICULE通过在原始和预先部署修补网络之间录制分歧来检测和隔离测试输入。然后培训Constcan以学习清洁验证和隔离输入之间的转换;即，它学会添加触发器来清洁验证图像。回顾验证图像以及其正确的标签用于进一步重新培训预修补程序，产生我们的最终防御。关于全面的后门攻击套件的实证评估表明，NNOCLICULE优于所有最先进的防御，以制定限制性假设，并且仅在特定的后门攻击上工作，或者在适应性攻击中失败。相比之下，NNICULICULE使得最小的假设并提供有效的防御，即使在现有防御因攻击者而导致其限制假设而导致的现有防御无效的情况下。

与令人印象深刻的进步触动了我们社会的各个方面，基于深度神经网络（DNN）的AI技术正在带来越来越多的安全问题。虽然在考试时间运行的攻击垄断了研究人员的初始关注，但是通过干扰培训过程来利用破坏DNN模型的可能性，代表了破坏训练过程的可能性，这是破坏AI技术的可靠性的进一步严重威胁。在后门攻击中，攻击者损坏了培训数据，以便在测试时间诱导错误的行为。然而，测试时间误差仅在存在与正确制作的输入样本对应的触发事件的情况下被激活。通过这种方式，损坏的网络继续正常输入的预期工作，并且只有当攻击者决定激活网络内隐藏的后门时，才会发生恶意行为。在过去几年中，后门攻击一直是强烈的研究活动的主题，重点是新的攻击阶段的发展，以及可能对策的提议。此概述文件的目标是审查发表的作品，直到现在，分类到目前为止提出的不同类型的攻击和防御。指导分析的分类基于攻击者对培训过程的控制量，以及防御者验证用于培训的数据的完整性，并监控DNN在培训和测试中的操作时间。因此，拟议的分析特别适合于参考他们在运营的应用方案的攻击和防御的强度和弱点。

A recent trojan attack on deep neural network (DNN) models is one insidious variant of data poisoning attacks. Trojan attacks exploit an effective backdoor created in a DNN model by leveraging the difficulty in interpretability of the learned model to misclassify any inputs signed with the attacker's chosen trojan trigger. Since the trojan trigger is a secret guarded and exploited by the attacker, detecting such trojan inputs is a challenge, especially at run-time when models are in active operation. This work builds STRong Intentional Perturbation (STRIP) based run-time trojan attack detection system and focuses on vision system. We intentionally perturb the incoming input, for instance by superimposing various image patterns, and observe the randomness of predicted classes for perturbed inputs from a given deployed model-malicious or benign. A low entropy in predicted classes violates the input-dependence property of a benign model and implies the presence of a malicious input-a characteristic of a trojaned input. The high efficacy of our method is validated through case studies on three popular and contrasting datasets: MNIST, CIFAR10 and GTSRB. We achieve an overall false acceptance rate (FAR) of less than 1%, given a preset false rejection rate (FRR) of 1%, for different types of triggers. Using CIFAR10 and GTSRB, we have empirically achieved result of 0% for both FRR and FAR. We have also evaluated STRIP robustness against a number of trojan attack variants and adaptive attacks.

典型的深神经网络（DNN）后门攻击基于输入中嵌入的触发因素。现有的不可察觉的触发因素在计算上昂贵或攻击成功率低。在本文中，我们提出了一个新的后门触发器，该扳机易于生成，不可察觉和高效。新的触发器是一个均匀生成的三维（3D）二进制图案，可以水平和/或垂直重复和镜像，并将其超级贴在三通道图像上，以训练后式DNN模型。新型触发器分散在整个图像中，对单个像素产生微弱的扰动，但共同拥有强大的识别模式来训练和激活DNN的后门。我们还通过分析表明，随着图像的分辨率提高，触发因素越来越有效。实验是使用MNIST，CIFAR-10和BTSR数据集上的RESNET-18和MLP模型进行的。在无遗象的方面，新触发的表现优于现有的触发器，例如Badnet，Trojaned NN和隐藏的后门。新的触发因素达到了几乎100％的攻击成功率，仅将分类准确性降低了不到0.7％-2.4％，并使最新的防御技术无效。

机器学习容易受到对抗操作的影响。先前的文献表明，在训练阶段，攻击者可以操纵数据和数据采样程序以控制模型行为。一个共同的攻击目标是种植后门，即迫使受害者模型学会识别只有对手知道的触发因素。在本文中，我们引入了一类新的后门攻击类，这些攻击隐藏在模型体系结构内，即在用于训练的功能的电感偏置中。这些后门很容易实现，例如，通过为其他人将在不知不觉中重复使用的后式模型体系结构发布开源代码。我们证明，模型架构后门代表了一个真正的威胁，与其他方法不同，可以从头开始进行完整的重新训练。我们将建筑后门背后的主要构建原理（例如输入和输出之间的链接）形式化，并描述对它们的一些可能的保护。我们评估了对不同尺度的计算机视觉基准测试的攻击，并证明在各种培训环境中，潜在的脆弱性无处不在。

计算能力和大型培训数据集的可用性增加，机器学习的成功助长了。假设它充分代表了在测试时遇到的数据，则使用培训数据来学习新模型或更新现有模型。这种假设受到中毒威胁的挑战，这种攻击会操纵训练数据，以损害模型在测试时的表现。尽管中毒已被认为是行业应用中的相关威胁，到目前为止，已经提出了各种不同的攻击和防御措施，但对该领域的完整系统化和批判性审查仍然缺失。在这项调查中，我们在机器学习中提供了中毒攻击和防御措施的全面系统化，审查了过去15年中该领域发表的100多篇论文。我们首先对当前的威胁模型和攻击进行分类，然后相应地组织现有防御。虽然我们主要关注计算机视觉应用程序，但我们认为我们的系统化还包括其他数据模式的最新攻击和防御。最后，我们讨论了中毒研究的现有资源，并阐明了当前的局限性和该研究领域的开放研究问题。

在对抗机器学习中，防止对深度学习系统的攻击的新防御能力在释放更强大的攻击后不久就会破坏。在这种情况下，法医工具可以通过追溯成功的根本原因来为现有防御措施提供宝贵的补充，并为缓解措施提供前进的途径，以防止将来采取类似的攻击。在本文中，我们描述了我们为开发用于深度神经网络毒物攻击的法医追溯工具的努力。我们提出了一种新型的迭代聚类和修剪解决方案，该解决方案修剪了“无辜”训练样本，直到所有剩余的是一组造成攻击的中毒数据。我们的方法群群训练样本基于它们对模型参数的影响，然后使用有效的数据解读方法来修剪无辜簇。我们从经验上证明了系统对三种类型的肮脏标签（后门）毒物攻击和三种类型的清洁标签毒药攻击的功效，这些毒物跨越了计算机视觉和恶意软件分类。我们的系统在所有攻击中都达到了98.4％的精度和96.8％的召回。我们还表明，我们的系统与专门攻击它的四种抗纤维法措施相对强大。

Backdoor attacks have emerged as one of the major security threats to deep learning models as they can easily control the model's test-time predictions by pre-injecting a backdoor trigger into the model at training time. While backdoor attacks have been extensively studied on images, few works have investigated the threat of backdoor attacks on time series data. To fill this gap, in this paper we present a novel generative approach for time series backdoor attacks against deep learning based time series classifiers. Backdoor attacks have two main goals: high stealthiness and high attack success rate. We find that, compared to images, it can be more challenging to achieve the two goals on time series. This is because time series have fewer input dimensions and lower degrees of freedom, making it hard to achieve a high attack success rate without compromising stealthiness. Our generative approach addresses this challenge by generating trigger patterns that are as realistic as real-time series patterns while achieving a high attack success rate without causing a significant drop in clean accuracy. We also show that our proposed attack is resistant to potential backdoor defenses. Furthermore, we propose a novel universal generator that can poison any type of time series with a single generator that allows universal attacks without the need to fine-tune the generative model for new time series datasets.

Dataset distillation has emerged as a prominent technique to improve data efficiency when training machine learning models. It encapsulates the knowledge from a large dataset into a smaller synthetic dataset. A model trained on this smaller distilled dataset can attain comparable performance to a model trained on the original training dataset. However, the existing dataset distillation techniques mainly aim at achieving the best trade-off between resource usage efficiency and model utility. The security risks stemming from them have not been explored. This study performs the first backdoor attack against the models trained on the data distilled by dataset distillation models in the image domain. Concretely, we inject triggers into the synthetic data during the distillation procedure rather than during the model training stage, where all previous attacks are performed. We propose two types of backdoor attacks, namely NAIVEATTACK and DOORPING. NAIVEATTACK simply adds triggers to the raw data at the initial distillation phase, while DOORPING iteratively updates the triggers during the entire distillation procedure. We conduct extensive evaluations on multiple datasets, architectures, and dataset distillation techniques. Empirical evaluation shows that NAIVEATTACK achieves decent attack success rate (ASR) scores in some cases, while DOORPING reaches higher ASR scores (close to 1.0) in all cases. Furthermore, we conduct a comprehensive ablation study to analyze the factors that may affect the attack performance. Finally, we evaluate multiple defense mechanisms against our backdoor attacks and show that our attacks can practically circumvent these defense mechanisms.

As a critical threat to deep neural networks (DNNs), backdoor attacks can be categorized into two types, i.e., source-agnostic backdoor attacks (SABAs) and source-specific backdoor attacks (SSBAs). Compared to traditional SABAs, SSBAs are more advanced in that they have superior stealthier in bypassing mainstream countermeasures that are effective against SABAs. Nonetheless, existing SSBAs suffer from two major limitations. First, they can hardly achieve a good trade-off between ASR (attack success rate) and FPR (false positive rate). Besides, they can be effectively detected by the state-of-the-art (SOTA) countermeasures (e.g., SCAn). To address the limitations above, we propose a new class of viable source-specific backdoor attacks, coined as CASSOCK. Our key insight is that trigger designs when creating poisoned data and cover data in SSBAs play a crucial role in demonstrating a viable source-specific attack, which has not been considered by existing SSBAs. With this insight, we focus on trigger transparency and content when crafting triggers for poisoned dataset where a sample has an attacker-targeted label and cover dataset where a sample has a ground-truth label. Specifically, we implement $CASSOCK_{Trans}$ and $CASSOCK_{Cont}$. While both they are orthogonal, they are complementary to each other, generating a more powerful attack, called $CASSOCK_{Comp}$, with further improved attack performance and stealthiness. We perform a comprehensive evaluation of the three $CASSOCK$-based attacks on four popular datasets and three SOTA defenses. Compared with a representative SSBA as a baseline ($SSBA_{Base}$), $CASSOCK$-based attacks have significantly advanced the attack performance, i.e., higher ASR and lower FPR with comparable CDA (clean data accuracy). Besides, $CASSOCK$-based attacks have effectively bypassed the SOTA defenses, and $SSBA_{Base}$ cannot.

解释性对于理解深神经网络（DNN）的内部工作至关重要，并且许多解释方法产生显着图，这些图突出了输入图像的一部分，这些图像对DNN的预测有了最大的影响。在本文中，我们设计了一种后门攻击，该攻击改变了网络为输入图像而改变的显着图，仅带有注入的触发器，而肉眼看不见，同时保持预测准确性。该攻击依赖于将中毒的数据注入训练数据集中。显着性图被合并到用于训练深层模型的目标函数的惩罚项中，其对模型训练的影响基于触发器的存在。我们设计了两种类型的攻击：有针对性的攻击，该攻击可以实施显着性图的特定修改和无靶向攻击的特定攻击，而当原始显着性图的顶部像素的重要性得分大大降低时。我们对针对各种深度学习体系结构的基于梯度和无梯度解释方法进行的后门攻击进行经验评估。我们表明，在部署不信任来源开发的深度学习模型时，我们的攻击构成了严重的安全威胁。最后，在补充中，我们证明了所提出的方法可以在倒置的设置中使用，在这种情况下，只有在存在触发器的情况下才能获得正确的显着性图（键），从而有效地使解释系统仅适用于选定的用户。

后门深度学习（DL）模型的行为通常在清洁输入上，但在触发器输入时不端行为，因为后门攻击者希望为DL模型部署构成严重后果。最先进的防御是限于特定的后门攻击（源无关攻击）或在该机器学习（ML）专业知识或昂贵的计算资源中不适用于源友好的攻击。这项工作观察到所有现有的后门攻击都具有不可避免的内在弱点，不可转换性，即触发器输入劫持劫持模型，但不能对另一个尚未植入同一后门的模型有效。通过此密钥观察，我们提出了不可转换性的反向检测（NTD）来识别运行时在运行时的模型欠测试（MUT）的触发输入。特定，NTD允许潜在的回溯静电预测输入的类别。同时，NTD利用特征提取器（FE）来提取输入的特征向量，并且从其预测类随机拾取的一组样本，然后比较FE潜在空间中的输入和样本之间的相似性。如果相似性低，则输入是对逆势触发输入;否则，良性。 FE是一个免费的预训练模型，私下从开放平台保留。随着FE和MUT来自不同来源，攻击者非常不可能将相同的后门插入其中两者。由于不可转换性，不能将突变处工作的触发效果转移到FE，使NTD对不同类型的后门攻击有效。我们在三个流行的定制任务中评估NTD，如面部识别，交通标志识别和一般动物分类，结果确认NDT具有高效率（低假验收率）和具有低检测延迟的可用性（低误报率）。

后门攻击已成为深度神经网络（DNN）的主要安全威胁。虽然现有的防御方法在检测或擦除后以后展示了有希望的结果，但仍然尚不清楚是否可以设计强大的培训方法，以防止后门触发器首先注入训练的模型。在本文中，我们介绍了\ emph {反后门学习}的概念，旨在培训\ emph {Clean}模型给出了后门中毒数据。我们将整体学习过程框架作为学习\ emph {clean}和\ emph {backdoor}部分的双重任务。从这种观点来看，我们确定了两个后门攻击的固有特征，因为他们的弱点2）后门任务与特定类（后门目标类）相关联。根据这两个弱点，我们提出了一般学习计划，反后门学习（ABL），在培训期间自动防止后门攻击。 ABL引入了标准培训的两级\ EMPH {梯度上升}机制，帮助分离早期训练阶段的后台示例，2）在后续训练阶段中断后门示例和目标类之间的相关性。通过对多个基准数据集的广泛实验，针对10个最先进的攻击，我们经验证明，后卫中毒数据上的ABL培训模型实现了与纯净清洁数据训练的相同性能。代码可用于\ url {https:/github.com/boylyg/abl}。

现代自动驾驶汽车采用最先进的DNN模型来解释传感器数据并感知环境。但是，DNN模型容易受到不同类型的对抗攻击的影响，这对车辆和乘客的安全性和安全性构成了重大风险。一个突出的威胁是后门攻击，对手可以通过中毒训练样本来妥协DNN模型。尽管已经大量精力致力于调查后门攻击对传统的计算机视觉任务，但很少探索其对自主驾驶场景的实用性和适用性，尤其是在物理世界中。在本文中，我们针对车道检测系统，该系统是许多自动驾驶任务，例如导航，车道切换的必不可少的模块。我们设计并实现了对此类系统的第一次物理后门攻击。我们的攻击是针对不同类型的车道检测算法的全面有效的。具体而言，我们引入了两种攻击方法（毒药和清洁量）来生成中毒样本。使用这些样品，训练有素的车道检测模型将被后门感染，并且可以通过公共物体（例如，交通锥）进行启动，以进行错误的检测，导致车辆从道路上或在相反的车道上行驶。对公共数据集和物理自动驾驶汽车的广泛评估表明，我们的后门攻击对各种防御解决方案都是有效，隐秘和强大的。我们的代码和实验视频可以在https://sites.google.com/view/lane-detection-attack/lda中找到。

后门（特洛伊木马）攻击正在对深度神经网络（DNN）产生威胁。每当来自任何源类的测试样本都嵌入后门图案时，DNN被攻击将预测到攻击者期望的目标类;在正确分类干净（无攻击）测试样本时。现有的后门防御在检测到DNN是攻击和逆向工程的“培训后”制度的反向工程方面取得了成功：防御者可以访问要检查的DNN和独立收集的小型清洁数据集，但是无法访问DNN的（可能中毒）培训集。然而，这些防御既不触发后门映射的行为也不抓住罪魁祸首，也不是在试验时间下减轻后门攻击。在本文中，我们提出了一个“飞行中的”防范反向攻击对图像分类的攻击，其中1）检测在试验时间时使用后门触发的使用; 2）Infers为检测到的触发器示例中的原始原点（源类）。我们防御的有效性是针对不同强大的后门攻击实验证明的。

最近的研究表明，深度神经网络（DNN）容易受到后门攻击的影响，后门攻击会导致DNN的恶意行为，当时特定的触发器附在输入图像上时。进一步证明，感染的DNN具有一系列通道，与正常通道相比，该通道对后门触发器更敏感。然后，将这些通道修剪可有效缓解后门行为。要定位这些通道，自然要考虑其Lipschitzness，这可以衡量他们对输入上最严重的扰动的敏感性。在这项工作中，我们介绍了一个名为Channel Lipschitz常数（CLC）的新颖概念，该概念定义为从输入图像到每个通道输出的映射的Lipschitz常数。然后，我们提供经验证据，以显示CLC（UCLC）上限与通道激活的触发激活变化之间的强相关性。由于可以从重量矩阵直接计算UCLC，因此我们可以以无数据的方式检测潜在的后门通道，并在感染的DNN上进行简单修剪以修复模型。提出的基于lipschitzness的通道修剪（CLP）方法非常快速，简单，无数据且可靠，可以选择修剪阈值。进行了广泛的实验来评估CLP的效率和有效性，CLP的效率和有效性也可以在主流防御方法中获得最新的结果。源代码可在https://github.com/rkteddy/channel-lipschitzness基于普通范围内获得。

最近的作品表明，深度学习模型容易受到后门中毒攻击的影响，在这些攻击中，这些攻击灌输了与外部触发模式或物体（例如贴纸，太阳镜等）的虚假相关性。我们发现这种外部触发信号是不必要的，因为可以使用基于旋转的图像转换轻松插入高效的后门。我们的方法通过旋转有限数量的对象并将其标记错误来构建中毒数据集；一旦接受过培训，受害者的模型将在运行时间推理期间做出不良的预测。它表现出明显的攻击成功率，同时通过有关图像分类和对象检测任务的全面实证研究来保持清洁绩效。此外，我们评估了标准数据增强技术和针对我们的攻击的四种不同的后门防御措施，发现它们都无法作为一致的缓解方法。正如我们在图像分类和对象检测应用程序中所示，我们的攻击只能在现实世界中轻松部署在现实世界中。总体而言，我们的工作突出了一个新的，简单的，物理上可实现的，高效的矢量，用于后门攻击。我们的视频演示可在https://youtu.be/6jif8wnx34m上找到。

Deep neural networks (DNNs) are vulnerable to a class of attacks called "backdoor attacks", which create an association between a backdoor trigger and a target label the attacker is interested in exploiting. A backdoored DNN performs well on clean test images, yet persistently predicts an attacker-defined label for any sample in the presence of the backdoor trigger. Although backdoor attacks have been extensively studied in the image domain, there are very few works that explore such attacks in the video domain, and they tend to conclude that image backdoor attacks are less effective in the video domain. In this work, we revisit the traditional backdoor threat model and incorporate additional video-related aspects to that model. We show that poisoned-label image backdoor attacks could be extended temporally in two ways, statically and dynamically, leading to highly effective attacks in the video domain. In addition, we explore natural video backdoors to highlight the seriousness of this vulnerability in the video domain. And, for the first time, we study multi-modal (audiovisual) backdoor attacks against video action recognition models, where we show that attacking a single modality is enough for achieving a high attack success rate.

In this paper, we present a simple yet surprisingly effective technique to induce "selective amnesia" on a backdoored model. Our approach, called SEAM, has been inspired by the problem of catastrophic forgetting (CF), a long standing issue in continual learning. Our idea is to retrain a given DNN model on randomly labeled clean data, to induce a CF on the model, leading to a sudden forget on both primary and backdoor tasks; then we recover the primary task by retraining the randomized model on correctly labeled clean data. We analyzed SEAM by modeling the unlearning process as continual learning and further approximating a DNN using Neural Tangent Kernel for measuring CF. Our analysis shows that our random-labeling approach actually maximizes the CF on an unknown backdoor in the absence of triggered inputs, and also preserves some feature extraction in the network to enable a fast revival of the primary task. We further evaluated SEAM on both image processing and Natural Language Processing tasks, under both data contamination and training manipulation attacks, over thousands of models either trained on popular image datasets or provided by the TrojAI competition. Our experiments show that SEAM vastly outperforms the state-of-the-art unlearning techniques, achieving a high Fidelity (measuring the gap between the accuracy of the primary task and that of the backdoor) within a few minutes (about 30 times faster than training a model from scratch using the MNIST dataset), with only a small amount of clean data (0.1% of training data for TrojAI models).