Search

Backdoor data poisoning, inserted within instruction examples used to fine-tune a foundation Large Language Model (LLM) for downstream tasks (\textit{e.g.,} sentiment prediction), is a serious security concern due to the evasive nature of such attacks. The poisoning is usually in the form of a (seemingly innocuous) trigger word or phrase inserted into a very small fraction of the fine-tuning samples from a target class. Such backdoor attacks can: alter response sentiment, violate censorship, over-refuse (invoke censorship for legitimate queries), inject false content, or trigger nonsense responses (hallucinations). In this work we investigate the efficacy of instruction fine-tuning backdoor attacks as attack "hyperparameters" are varied under a variety of scenarios, considering: the trigger location in the poisoned examples; robustness to change in the trigger location, partial triggers, and synonym substitutions at test time; attack transfer from one (fine-tuning) domain to a related test domain; and clean-label vs. dirty-label poisoning. Based on our observations, we propose and evaluate two defenses against these attacks: i) a \textit{during-fine-tuning defense} based on word-frequency counts that assumes the (possibly poisoned) fine-tuning dataset is available and identifies the backdoor trigger tokens; and ii) a \textit{post-fine-tuning defense} based on downstream clean fine-tuning of the backdoored LLM with a small defense dataset. Finally, we provide a brief survey of related work on backdoor attacks and defenses., Comment: Under review

A variety of defenses have been proposed against backdoors attacks on deep neural network (DNN) classifiers. Universal methods seek to reliably detect and/or mitigate backdoors irrespective of the incorporation mechanism used by the attacker, while reverse-engineering methods often explicitly assume one. In this paper, we describe a new detector that: relies on internal feature map of the defended DNN to detect and reverse-engineer the backdoor and identify its target class; can operate post-training (without access to the training dataset); is highly effective for various incorporation mechanisms (i.e., is universal); and which has low computational overhead and so is scalable. Our detection approach is evaluated for different attacks on benchmark CIFAR-10 and CIFAR-100 image classifiers.

Deep neural networks (DNNs) have achieved tremendous success in various applications including video action recognition, yet remain vulnerable to backdoor attacks (Trojans). The backdoor-compromised model will mis-classify to the target class chosen by the attacker when a test instance (from a non-target class) is embedded with a specific trigger, while maintaining high accuracy on attack-free instances. Although there are extensive studies on backdoor attacks against image data, the susceptibility of video-based systems under backdoor attacks remains largely unexplored. Current studies are direct extensions of approaches proposed for image data, e.g., the triggers are independently embedded within the frames, which tend to be detectable by existing defenses. In this paper, we introduce a simple yet effective backdoor attack against video data. Our proposed attack, adding perturbations in a transformed domain, plants an imperceptible, temporally distributed trigger across the video frames, and is shown to be resilient to existing defensive strategies. The effectiveness of the proposed attack is demonstrated by extensive experiments with various well-known models on two video recognition benchmarks, UCF101 and HMDB51, and a sign language recognition benchmark, Greek Sign Language (GSL) dataset. We delve into the impact of several influential factors on our proposed attack and identify an intriguing effect termed "collateral damage" through extensive studies., Comment: accepted by AAAI 2024

Backdoor (Trojan) attacks are an important type of adversarial exploit against deep neural networks (DNNs), wherein a test instance is (mis)classified to the attacker's target class whenever the attacker's backdoor trigger is present. In this paper, we reveal and analyze an important property of backdoor attacks: a successful attack causes an alteration in the distribution of internal layer activations for backdoor-trigger instances, compared to that for clean instances. Even more importantly, we find that instances with the backdoor trigger will be correctly classified to their original source classes if this distribution alteration is corrected. Based on our observations, we propose an efficient and effective method that achieves post-training backdoor mitigation by correcting the distribution alteration using reverse-engineered triggers. Notably, our method does not change any trainable parameters of the DNN, but achieves generally better mitigation performance than existing methods that do require intensive DNN parameter tuning. It also efficiently detects test instances with the trigger, which may help to catch adversarial entities in the act of exploiting the backdoor.

Deep neural networks are vulnerable to backdoor attacks (Trojans), where an attacker poisons the training set with backdoor triggers so that the neural network learns to classify test-time triggers to the attacker's designated target class. Recent work shows that backdoor poisoning induces over-fitting (abnormally large activations) in the attacked model, which motivates a general, post-training clipping method for backdoor mitigation, i.e., with bounds on internal-layer activations learned using a small set of clean samples. We devise a new such approach, choosing the activation bounds to explicitly limit classification margins. This method gives superior performance against peer methods for CIFAR-10 image classification. We also show that this method has strong robustness against adaptive attacks, X2X attacks, and on different datasets. Finally, we demonstrate a method extension for test-time detection and correction based on the output differences between the original and activation-bounded networks. The code of our method is online available.

Backdoor attacks are an important type of adversarial threat against deep neural network classifiers, wherein test samples from one or more source classes will be (mis)classified to the attacker's target class when a backdoor pattern is embedded. In this paper, we focus on the post-training backdoor defense scenario commonly considered in the literature, where the defender aims to detect whether a trained classifier was backdoor-attacked without any access to the training set. Many post-training detectors are designed to detect attacks that use either one or a few specific backdoor embedding functions (e.g., patch-replacement or additive attacks). These detectors may fail when the backdoor embedding function used by the attacker (unknown to the defender) is different from the backdoor embedding function assumed by the defender. In contrast, we propose a post-training defense that detects backdoor attacks with arbitrary types of backdoor embeddings, without making any assumptions about the backdoor embedding type. Our detector leverages the influence of the backdoor attack, independent of the backdoor embedding mechanism, on the landscape of the classifier's outputs prior to the softmax layer. For each class, a maximum margin statistic is estimated. Detection inference is then performed by applying an unsupervised anomaly detector to these statistics. Thus, our detector does not need any legitimate clean samples, and can efficiently detect backdoor attacks with arbitrary numbers of source classes. These advantages over several state-of-the-art methods are demonstrated on four datasets, for three different types of backdoor patterns, and for a variety of attack configurations. Finally, we propose a novel, general approach for backdoor mitigation once a detection is made. The mitigation approach was the runner-up at the first IEEE Trojan Removal Competition. The code is online available.

Backdoor attacks (BAs) are an emerging threat to deep neural network classifiers. A victim classifier will predict to an attacker-desired target class whenever a test sample is embedded with the same backdoor pattern (BP) that was used to poison the classifier's training set. Detecting whether a classifier is backdoor attacked is not easy in practice, especially when the defender is, e.g., a downstream user without access to the classifier's training set. This challenge is addressed here by a reverse-engineering defense (RED), which has been shown to yield state-of-the-art performance in several domains. However, existing REDs are not applicable when there are only {\it two classes} or when {\it multiple attacks} are present. These scenarios are first studied in the current paper, under the practical constraints that the defender neither has access to the classifier's training set nor to supervision from clean reference classifiers trained for the same domain. We propose a detection framework based on BP reverse-engineering and a novel {\it expected transferability} (ET) statistic. We show that our ET statistic is effective {\it using the same detection threshold}, irrespective of the classification domain, the attack configuration, and the BP reverse-engineering algorithm that is used. The excellent performance of our method is demonstrated on six benchmark datasets. Notably, our detection framework is also applicable to multi-class scenarios with multiple attacks. Code is available at https://github.com/zhenxianglance/2ClassBADetection., Comment: Accepted to ICLR2022

Backdoor (Trojan) attacks are emerging threats against deep neural networks (DNN). A DNN being attacked will predict to an attacker-desired target class whenever a test sample from any source class is embedded with a backdoor pattern; while correctly classifying clean (attack-free) test samples. Existing backdoor defenses have shown success in detecting whether a DNN is attacked and in reverse-engineering the backdoor pattern in a "post-training" regime: the defender has access to the DNN to be inspected and a small, clean dataset collected independently, but has no access to the (possibly poisoned) training set of the DNN. However, these defenses neither catch culprits in the act of triggering the backdoor mapping, nor mitigate the backdoor attack at test-time. In this paper, we propose an "in-flight" defense against backdoor attacks on image classification that 1) detects use of a backdoor trigger at test-time; and 2) infers the class of origin (source class) for a detected trigger example. The effectiveness of our defense is demonstrated experimentally against different strong backdoor attacks.

Backdoor attacks (BA) are an emerging threat to deep neural network classifiers. A classifier being attacked will predict to the attacker's target class when a test sample from a source class is embedded with the backdoor pattern (BP). Recently, the first BA against point cloud (PC) classifiers was proposed, creating new threats to many important applications including autonomous driving. Such PC BAs are not detectable by existing BA defenses due to their special BP embedding mechanism. In this paper, we propose a reverse-engineering defense that infers whether a PC classifier is backdoor attacked, without access to its training set or to any clean classifiers for reference. The effectiveness of our defense is demonstrated on the benchmark ModeNet40 dataset for PCs.

We demonstrate a backdoor attack on a deep neural network used for regression. The backdoor attack is localized based on training-set data poisoning wherein the mislabeled samples are surrounded by correctly labeled ones. We demonstrate how such localization is necessary for attack success. We also study the performance of a backdoor defense using gradient-based discovery of local error maximizers. Local error maximizers which are associated with significant (interpolation) error, and are proximal to many training samples, are suspicious. This method is also used to accurately train for deep regression in the first place by active (deep) learning leveraging an "oracle" capable of providing real-valued supervision (a regression target) for samples. Such oracles, including traditional numerical solvers of PDEs or SDEs using finite difference or Monte Carlo approximations, are far more computationally costly compared to deep regression.

Data Poisoning (DP) is an effective attack that causes trained classifiers to misclassify their inputs. DP attacks significantly degrade a classifier's accuracy by covertly injecting attack samples into the training set. Broadly applicable to different classifier structures, without strong assumptions about the attacker, an {\it unsupervised} Bayesian Information Criterion (BIC)-based mixture model defense against "error generic" DP attacks is herein proposed that: 1) addresses the most challenging {\it embedded} DP scenario wherein, if DP is present, the poisoned samples are an {\it a priori} unknown subset of the training set, and with no clean validation set available; 2) applies a mixture model both to well-fit potentially multi-modal class distributions and to capture poisoned samples within a small subset of the mixture components; 3) jointly identifies poisoned components and samples by minimizing the BIC cost defined over the whole training set, with the identified poisoned data removed prior to classifier training. Our experimental results, for various classifier structures and benchmark datasets, demonstrate the effectiveness and universality of our defense under strong DP attacks, as well as its superiority over other works.

Vulnerability of 3D point cloud (PC) classifiers has become a grave concern due to the popularity of 3D sensors in safety-critical applications. Existing adversarial attacks against 3D PC classifiers are all test-time evasion (TTE) attacks that aim to induce test-time misclassifications using knowledge of the classifier. But since the victim classifier is usually not accessible to the attacker, the threat is largely diminished in practice, as PC TTEs typically have poor transferability. Here, we propose the first backdoor attack (BA) against PC classifiers. Originally proposed for images, BAs poison the victim classifier's training set so that the classifier learns to decide to the attacker's target class whenever the attacker's backdoor pattern is present in a given input sample. Significantly, BAs do not require knowledge of the victim classifier. Different from image BAs, we propose to insert a cluster of points into a PC as a robust backdoor pattern customized for 3D PCs. Such clusters are also consistent with a physical attack (i.e., with a captured object in a scene). We optimize the cluster's location using an independently trained surrogate classifier and choose the cluster's local geometry to evade possible PC preprocessing and PC anomaly detectors (ADs). Experimentally, our BA achieves a uniformly high success rate (> 87%) and shows evasiveness against state-of-the-art PC ADs.

Recently, a special type of data poisoning (DP) attack targeting Deep Neural Network (DNN) classifiers, known as a backdoor, was proposed. These attacks do not seek to degrade classification accuracy, but rather to have the classifier learn to classify to a target class whenever the backdoor pattern is present in a test example. Launching backdoor attacks does not require knowledge of the classifier or its training process - it only needs the ability to poison the training set with (a sufficient number of) exemplars containing a sufficiently strong backdoor pattern (labeled with the target class). Here we address post-training detection of backdoor attacks in DNN image classifiers, seldom considered in existing works, wherein the defender does not have access to the poisoned training set, but only to the trained classifier itself, as well as to clean examples from the classification domain. This is an important scenario because a trained classifier may be the basis of e.g. a phone app that will be shared with many users. Detecting backdoors post-training may thus reveal a widespread attack. We propose a purely unsupervised anomaly detection (AD) defense against imperceptible backdoor attacks that: i) detects whether the trained DNN has been backdoor-attacked; ii) infers the source and target classes involved in a detected attack; iii) we even demonstrate it is possible to accurately estimate the backdoor pattern. We test our AD approach, in comparison with alternative defenses, for several backdoor patterns, data sets, and attack settings and demonstrate its favorability. Our defense essentially requires setting a single hyperparameter (the detection threshold), which can e.g. be chosen to fix the system's false positive rate.

There is great potential for damage from adversarial learning (AL) attacks on machine-learning based systems. In this paper, we provide a contemporary survey of AL, focused particularly on defenses against attacks on statistical classifiers. After introducing relevant terminology and the goals and range of possible knowledge of both attackers and defenders, we survey recent work on test-time evasion (TTE), data poisoning (DP), and reverse engineering (RE) attacks and particularly defenses against same. In so doing, we distinguish robust classification from anomaly detection (AD), unsupervised from supervised, and statistical hypothesis-based defenses from ones that do not have an explicit null (no attack) hypothesis; we identify the hyperparameters a particular method requires, its computational complexity, as well as the performance measures on which it was evaluated and the obtained quality. We then dig deeper, providing novel insights that challenge conventional AL wisdom and that target unresolved issues, including: 1) robust classification versus AD as a defense strategy; 2) the belief that attack success increases with attack strength, which ignores susceptibility to AD; 3) small perturbations for test-time evasion attacks: a fallacy or a requirement?; 4) validity of the universal assumption that a TTE attacker knows the ground-truth class for the example to be attacked; 5) black, grey, or white box attacks as the standard for defense evaluation; 6) susceptibility of query-based RE to an AD defense. We also discuss attacks on the privacy of training data. We then present benchmark comparisons of several defenses against TTE, RE, and backdoor DP attacks on images. The paper concludes with a discussion of future work.

Naive Bayes spam filters are highly susceptible to data poisoning attacks. Here, known spam sources/blacklisted IPs exploit the fact that their received emails will be treated as (ground truth) labeled spam examples, and used for classifier training (or re-training). The attacking source thus generates emails that will skew the spam model, potentially resulting in great degradation in classifier accuracy. Such attacks are successful mainly because of the poor representation power of the naive Bayes (NB) model, with only a single (component) density to represent spam (plus a possible attack). We propose a defense based on the use of a mixture of NB models. We demonstrate that the learned mixture almost completely isolates the attack in a second NB component, with the original spam component essentially unchanged by the attack. Our approach addresses both the scenario where the classifier is being re-trained in light of new data and, significantly, the more challenging scenario where the attack is embedded in the original spam training set. Even for weak attack strengths, BIC-based model order selection chooses a two-component solution, which invokes the mixture-based defense. Promising results are presented on the TREC 2005 spam corpus.

A significant threat to the recent, wide deployment of machine learning-based systems, including deep neural networks (DNNs), is adversarial learning attacks. We analyze possible test-time evasion-attack mechanisms and show that, in some important cases, when the image has been attacked, correctly classifying it has no utility: i) when the image to be attacked is (even arbitrarily) selected from the attacker's cache; ii) when the sole recipient of the classifier's decision is the attacker. Moreover, in some application domains and scenarios it is highly actionable to detect the attack irrespective of correctly classifying in the face of it (with classification still performed if no attack is detected). We hypothesize that, even if human-imperceptible, adversarial perturbations are machine-detectable. We propose a purely unsupervised anomaly detector (AD) that, unlike previous works: i) models the joint density of a deep layer using highly suitable null hypothesis density models (matched in particular to the non- negative support for RELU layers); ii) exploits multiple DNN layers; iii) leverages a "source" and "destination" class concept, source class uncertainty, the class confusion matrix, and DNN weight information in constructing a novel decision statistic grounded in the Kullback-Leibler divergence. Tested on MNIST and CIFAR-10 image databases under three prominent attack strategies, our approach outperforms previous detection methods, achieving strong ROC AUC detection accuracy on two attacks and better accuracy than recently reported for a variety of methods on the strongest (CW) attack. We also evaluate a fully white box attack on our system. Finally, we evaluate other important performance measures, such as classification accuracy, versus detection rate and attack strength.

We consider a cloud based multiserver system consisting of a set of replica application servers behind a set of proxy (indirection) servers which interact directly with clients over the Internet. We study a proactive moving-target defense to thwart a DDoS attacker's reconnaissance phase and consequently reduce the attack's impact. The defense is effectively a moving-target (motag) technique in which the proxies dynamically change. The system is evaluated using an AWS prototype of HTTP redirection and by numerical evaluations of an adversarial coupon-collector mathematical model, the latter allowing larger-scale extrapolations.

This papers consists of two parts. The first is a critical review of prior art on adversarial learning, identifying some significant limitations of previous works. The second part is an experimental study considering adversarial active learning and an investigation of the efficacy of a mixed sample selection strategy for combating an adversary who attempts to disrupt the classifier learning.

We consider a cloud based multiserver system, that may be cloud based, consisting of a set of replica application servers behind a set of proxy (indirection) servers which interact directly with clients over the Internet. We address cloud-side proactive and reactive defenses to combat DDoS attacks that may target this system. DDoS attacks are endemic with some notable attacks occurring just this past fall. Volumetric attacks may target proxies while "low volume" attacks may target replicas. After reviewing existing and proposed defenses, such as changing proxy IP addresses (a "moving target" technique to combat the reconnaissance phase of the botnet) and fission of overloaded servers, we focus on evaluation of defenses based on shuffling client-to-server assignments that can be both proactive and reactive to a DDoS attack. Our evaluations are based on a binomial distribution model that well agrees with simulations and preliminary experiments on a prototype that is also described.

In a variety of applications, one desires to detect groups of anomalous data samples, with a group potentially manifesting its atypicality (relative to a reference model) on a low-dimensional subset of the full measured set of features. Samples may only be weakly atypical individually, whereas they may be strongly atypical when considered jointly. What makes this group anomaly detection problem quite challenging is that it is a priori unknown which subset of features jointly manifests a particular group of anomalies. Moreover, it is unknown how many anomalous groups are present in a given data batch. In this work, we develop a group anomaly detection (GAD) scheme to identify the subset of samples and subset of features that jointly specify an anomalous cluster. We apply our approach to network intrusion detection to detect BotNet and peer-to-peer flow clusters. Unlike previous studies, our approach captures and exploits statistical dependencies that may exist between the measured features. Experiments on real world network traffic data demonstrate the advantage of our proposed system, and highlight the importance of exploiting feature dependency structure, compared to the feature (or test) independence assumption made in previous studies.