预训练模型(PTM)已被广泛用于各种下游任务。 PTM的参数分布在Internet上,可能会遭受后门攻击。在这项工作中,我们演示了PTMS的普遍脆弱性,在该工作中,可以通过任意下游任务中的后门攻击轻松控制PTMS。具体而言,攻击者可以添加一个简单的预训练任务,该任务将触发实例的输出表示限制为预定义的向量,即神经元级后门攻击(NEUBA)。如果在微调过程中未消除后门功能,则触发器可以通过预定义的矢量预测固定标签。在自然语言处理(NLP)和计算机视觉(CV)的实验中,我们表明Neuba绝对可以控制触发实例的预测,而无需了解下游任务。最后,我们将几种防御方法应用于Neuba,并发现模型修剪是通过排除后门神经元来抵抗Neuba的有希望的方向。我们的发现听起来是红色警报,用于广泛使用PTM。我们的源代码和模型可在\ url {https://github.com/thunlp/neuba}上获得。
translated by 谷歌翻译
文本后门攻击是对NLP系统的实际威胁。通过在训练阶段注入后门,对手可以通过预定义的触发器控制模型预测。由于已经提出了各种攻击和防御模型,因此进行严格的评估至关重要。但是,我们在以前的后门学习评估中重点介绍了两个问题:(1)忽略了现实世界情景(例如释放中毒的数据集或模型)之间的差异,我们认为每种情况都有其自身的限制和关注点,因此需要特定的评估。协议; (2)评估指标仅考虑攻击是否可以翻转模型对中毒样品的预测并保留对良性样品的表演,但是忽略了中毒样品也应该是隐秘和语义上的。为了解决这些问题,我们将现有作品分为三种实际情况,在这种情况下,攻击者分别释放数据集,预培训模型和微调模型,然后讨论其独特的评估方法。关于指标,为了完全评估中毒样本,我们使用语法误差增加和隐形性差异以及有效性的文本相似性。对框架进行正式化后,我们开发了一个开源工具包openbackdoor,以促进文本后门学习的实现和评估。使用此工具包,我们在建议的范式下进行基准攻击和防御模型进行广泛的实验。为了促进针对中毒数据集的不充分的防御能力,我们进一步提出了Cube,这是一个简单而强大的基于聚类的防御基线。我们希望我们的框架和基准可以作为未来模型开发和评估的基石。
translated by 谷歌翻译
Backdoor attacks represent one of the major threats to machine learning models. Various efforts have been made to mitigate backdoors. However, existing defenses have become increasingly complex and often require high computational resources or may also jeopardize models' utility. In this work, we show that fine-tuning, one of the most common and easy-to-adopt machine learning training operations, can effectively remove backdoors from machine learning models while maintaining high model utility. Extensive experiments over three machine learning paradigms show that fine-tuning and our newly proposed super-fine-tuning achieve strong defense performance. Furthermore, we coin a new term, namely backdoor sequela, to measure the changes in model vulnerabilities to other attacks before and after the backdoor has been removed. Empirical evaluation shows that, compared to other defense methods, super-fine-tuning leaves limited backdoor sequela. We hope our results can help machine learning model owners better protect their models from backdoor threats. Also, it calls for the design of more advanced attacks in order to comprehensively assess machine learning models' backdoor vulnerabilities.
translated by 谷歌翻译
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.
translated by 谷歌翻译
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。
translated by 谷歌翻译
最近,已经表明,自然语言处理(NLP)模型容易受到一种称为后门攻击的安全威胁,它利用“后门触发器”范例误导模型。最威胁的后门攻击是隐身的后门,它将触发器定义为文本样式或句法。虽然他们已经取得了令人难以置信的高攻击成功率(ASR),但我们发现为ASR的主要因素贡献不是“后门触发”范式。因此,当作为后门攻击分类时,这些隐身后门攻击的能力大得多。因此,为了评估后门攻击的真正攻击力,我们提出了一种称为攻击成功率差异(ASRD)的新度量,从而测量干净状态和毒药状态模型之间的ASR差异。此外,由于对抗隐蔽的后门攻击的防御,我们提出了触发破坏者,包括两个太简单的技巧,可以有效地防御隐秘的后门攻击。关于文本分类任务的实验表明,我们的方法比对隐身后门攻击的最先进的防御方法实现了更好的性能。
translated by 谷歌翻译
We conduct a systematic study of backdoor vulnerabilities in normally trained Deep Learning models. They are as dangerous as backdoors injected by data poisoning because both can be equally exploited. We leverage 20 different types of injected backdoor attacks in the literature as the guidance and study their correspondences in normally trained models, which we call natural backdoor vulnerabilities. We find that natural backdoors are widely existing, with most injected backdoor attacks having natural correspondences. We categorize these natural backdoors and propose a general detection framework. It finds 315 natural backdoors in the 56 normally trained models downloaded from the Internet, covering all the different categories, while existing scanners designed for injected backdoors can at most detect 65 backdoors. We also study the root causes and defense of natural backdoors.
translated by 谷歌翻译
后门攻击已被证明是对深度学习模型的严重安全威胁,并且检测给定模型是否已成为后门成为至关重要的任务。现有的防御措施主要建立在观察到后门触发器通常尺寸很小或仅影响几个神经元激活的观察结果。但是,在许多情况下,尤其是对于高级后门攻击,违反了上述观察结果,阻碍了现有防御的性能和适用性。在本文中,我们提出了基于新观察的后门防御范围。也就是说,有效的后门攻击通常需要对中毒训练样本的高预测置信度,以确保训练有素的模型具有很高的可能性。基于此观察结果,Dtinspector首先学习一个可以改变最高信心数据的预测的补丁,然后通过检查在低信心数据上应用学习补丁后检查预测变化的比率来决定后门的存在。对五次后门攻击,四个数据集和三种高级攻击类型的广泛评估证明了拟议防御的有效性。
translated by 谷歌翻译
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}。
translated by 谷歌翻译
后门攻击对NLP模型构成了新的威胁。在后门攻击中构建中毒数据的标准策略是将触发器(例如,稀有字)插入所选句子,并将原始标签更改为目标标签。该策略具有从触发器和标签视角轻松检测到的严重缺陷:注入的触发器,通常是一种罕见的单词,导致异常的自然语言表达,因此可以通过防御模型容易地检测到异常的自然语言表达;改变的目标标签会导致误报标记的示例,因此可以通过手动检查容易地检测到。要处理此问题,请在本文中,我们提出了一种新的策略来执行不需要外部触发的文本后门攻击,并且中毒样品被正确标记。拟议策略的核心思想是构建清洁标记的例子,其标签是正确的,但可以导致测试标签在与培训集合融合时的变化。为了产生中毒清洁标记的例子,我们提出了一种基于遗传算法的句子生成模型,以满足文本数据的不可微差特性。广泛的实验表明,拟议的攻击策略不仅有效,而且更重要的是,由于其令人触发和清洁的性质,难以防御。我们的工作标志着在NLP中开发令人触发的攻击策略的第一步。
translated by 谷歌翻译
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.
translated by 谷歌翻译
已知深层神经网络(DNN)容易受到后门攻击和对抗攻击的影响。在文献中,这两种攻击通常被视为明显的问题并分别解决,因为它们分别属于训练时间和推理时间攻击。但是,在本文中,我们发现它们之间有一个有趣的联系:对于具有后门种植的模型,我们观察到其对抗性示例具有与触发样品相似的行为,即都激活了同一DNN神经元的子集。这表明将后门种植到模型中会严重影响模型的对抗性例子。基于这一观察结果,我们设计了一种新的对抗性微调(AFT)算法,以防止后门攻击。我们从经验上表明,在5次最先进的后门攻击中,我们的船尾可以有效地擦除后门触发器,而无需在干净的样品上明显的性能降解,并显着优于现有的防御方法。
translated by 谷歌翻译
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).
translated by 谷歌翻译
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。
translated by 谷歌翻译
后门学习是研究深神经网络(DNNS)脆弱性的一个新兴而重要的话题。在快速武器竞赛的地位上,正在连续或同时提出许多开创性的后门攻击和防御方法。但是,我们发现对新方法的评估通常是不可思议的,以验证其主张和实际绩效,这主要是由于快速发展,不同的环境以及实施和可重复性的困难。没有彻底的评估和比较,很难跟踪当前的进度并设计文献的未来发展路线图。为了减轻这一困境,我们建立了一个名为Backdoorbench的后门学习的全面基准。它由一个可扩展的基于模块化的代码库(当前包括8个最先进(SOTA)攻击和9种SOTA防御算法的实现),以及完整的后门学习的标准化协议。我们还基于5个模型和4个数据集,对9个防御措施的每对8次攻击进行全面评估,总共8,000对评估。我们从不同的角度进一步介绍了对这8,000次评估的不同角度,研究了对国防算法,中毒比率,模型和数据集对后门学习的影响。 \ url {https://backdoorbench.github.io}公开获得了Backdoorbench的所有代码和评估。
translated by 谷歌翻译
最近,变压器架构已经证明了其在自然语言处理(NLP)和计算机视觉(CV)任务中的重要性。虽然已知其他网络模型容易受到后门攻击的影响,但是在模型中嵌入触发器并在呈现触发器时控制模型行为,众所周知,这种攻击是否仍然在变压器模型上仍然有效,如果是的话,是否有效它可以以更具成本效益的方式完成。在本文中,我们提出DBIA,一种对CV导向的变压器网络的一种新型无数据响应攻击,利用变压器的固有注意机制来产生触发器并使用中毒代理数据集注入后门。我们在两个主流图像分类任务中基于三个基准变压器,即Vit,Deit和Swin变压器进行了广泛的实验..,Cifar10和ImageNet。评估结果表明,消耗较少的资源,我们的方法可以嵌入高层的成功率和对受害者变压器性能的低影响。我们的代码可在https://anonmous.4open.science/r/dbia-825d获得。
translated by 谷歌翻译
后门攻击是对深度神经网络(DNN)的一种紧急培训时间威胁。它们可以操纵DNN的输出并具有高度思虑。在自然语言处理领域,已经提出了一些攻击方法,并在多个流行型号上实现了非常高的攻击成功率。尽管如此,很少有关于捍卫文本后门攻击的研究。在本文中,我们提出了一个简单且有效的文本后门防御,名为洋葱,这是基于异常字检测,并据我们所知,是可以处理所有文本后门攻击情况的第一种方法。实验证明了我们模型在捍卫Bilstm和BERT的措施与五种不同的后门攻击的有效性。本文的所有代码和数据都可以在https://github.com/thunlp/onion获得。
translated by 谷歌翻译
视觉变压器(VITS)具有与卷积神经网络相比,具有较小的感应偏置的根本不同的结构。随着绩效的提高,VIT的安全性和鲁棒性也非常重要。与许多最近利用VIT反对对抗性例子的鲁棒性的作品相反,本文调查了代表性的病因攻击,即后门。我们首先检查了VIT对各种后门攻击的脆弱性,发现VIT也很容易受到现有攻击的影响。但是,我们观察到,VIT的清洁数据准确性和后门攻击成功率在位置编码之前对补丁转换做出了明显的反应。然后,根据这一发现,我们为VIT提出了一种通过补丁处理来捍卫基于补丁的触发后门攻击的有效方法。在包括CIFAR10,GTSRB和Tinyimagenet在内的几个基准数据集上评估了这些表演,这些数据表明,该拟议的新颖防御在减轻VIT的后门攻击方面非常成功。据我们所知,本文提出了第一个防御性策略,该策略利用了反对后门攻击的VIT的独特特征。
translated by 谷歌翻译
神经网络的不透明度导致其脆弱性发生后门攻击,其中触发了感染神经元的隐藏注意力,以覆盖对攻击者选择的神经元的正常预测。在本文中,我们提出了一种新型的后门防御方法,以标记和净化后门神经网络中受感染的神经元。具体来说,我们首先定义了一个名为良性显着性的新指标。通过将一阶梯度组合以保持神经元之间的连接,良性显着性可以鉴定出比后门防御中常用度量的高精度的感染神经元。然后,提出了一种新的自适应正则化(AR)机制,以通过微调来帮助净化这些被鉴定的感染神经元。由于能够适应不同参数幅度的能力,与神经元纯化中的共同正则化机制相比,AR可以提供更快,更稳定的收敛性。广泛的实验结果表明,我们的方法可以消除具有可忽略的性能降解的神经网络中的后门。
translated by 谷歌翻译
This paper asks the intriguing question: is it possible to exploit neural architecture search (NAS) as a new attack vector to launch previously improbable attacks? Specifically, we present EVAS, a new attack that leverages NAS to find neural architectures with inherent backdoors and exploits such vulnerability using input-aware triggers. Compared with existing attacks, EVAS demonstrates many interesting properties: (i) it does not require polluting training data or perturbing model parameters; (ii) it is agnostic to downstream fine-tuning or even re-training from scratch; (iii) it naturally evades defenses that rely on inspecting model parameters or training data. With extensive evaluation on benchmark datasets, we show that EVAS features high evasiveness, transferability, and robustness, thereby expanding the adversary's design spectrum. We further characterize the mechanisms underlying EVAS, which are possibly explainable by architecture-level ``shortcuts'' that recognize trigger patterns. This work raises concerns about the current practice of NAS and points to potential directions to develop effective countermeasures.
translated by 谷歌翻译