Machine learning (ML) suffers from a persistent and critical flaw: adversarial examples. Many new forms of adversarial example attacks have been invented and many narrow defenses have been proposed. Unfortunately, no defensive approach can withstand current attacks. We hypothesize that ML model robustness can be improved with approaches that delineate the data-point-sparse latent space between data-dense regions of a model’s classification space as a barrier class. We introduce one such defense, PadNet, that builds a barrier class using a combination of training samples that mix multiple classes together. It leverages this barrier class to separate decision boundaries between benign classes with regions of padding. PadNet then implements a gradient regularization strategy that penalizes gradients in the direction of the barrier class, causing the decision boundary to draw tighter around training samples increasing boundary thickness between classes. We evaluate PadNet against a sampling of the most effective state-of-the-art attacks, demonstrating that it offers significant robustness and reliability compared to current defenses. We also test it against adaptive attacks and find that PadNet remains robust against them.

Continual Learning (CL) for malware classification tackles the rapidly evolving nature of malware threats and the frequent emergence of new types. Generative Replay (GR)-based CL systems utilize a generative model to produce synthetic versions of past data, which are then combined with new data to retrain the primary model. Traditional machine learning techniques in this domain often struggle with catastrophic forgetting, where a model's performance on old data degrades over time. In this paper, we introduce a GR-based CL system that employs Generative Adversarial Networks (GANs) with feature matching loss to generate high-quality malware samples. Additionally, we implement innovative selection schemes for replay samples based on the model’s hidden representations. Our comprehensive evaluation across Windows and Android malware datasets in a class-incremental learning scenario -- where new classes are introduced continuously over multiple tasks -- demonstrates substantial performance improvements over previous methods. For example, our system achieves an average accuracy of 55% on Windows malware samples, significantly outperforming other GR-based models by 28%. This study provides practical insights for advancing GR-based malware classification systems.

Recent website fingerprinting attacks have been shown to achieve very high performance against traffic through Tor. These attacks allow an adversary to deduce the website a Tor user has visited by simply eavesdropping on the encrypted communication. This has consequently motivated the development of many defense strategies that obfuscate traffic through the addition of dummy packets and/or delays. The efficacy and practicality of many of these recent proposals have yet to be scrutinized in detail. In this study, we re-evaluate nine recent defense proposals that claim to provide adequate security with low-overheads using the latest Deep Learning-based attacks. Furthermore, we assess the feasibility of implementing these defenses within the current confines of Tor. To this end, we additionally provide the first on-network implementation of the DynaFlow defense to better assess its real-world utility.

End-to-end flow correlation attacks are among the oldest known attacks on low-latency anonymity networks, and are treated as a core primitive for traffic analysis of Tor. However, despite recent work showing that individual flows can be correlated with high accuracy, the impact of even these state-of-the-art attacks is questionable due to a central drawback: their pairwise nature, requiring comparison between N2pairs of flows to deanonymize N users. This results in a combinatorial explosion in computational requirements and an asymptotically declining base rate, leading to either high numbers of false positives or vanishingly small rates of successful correlation. In this paper, we introduce a novel flow correlation attack, DeepCoFFEA, that combines two ideas to overcome these drawbacks. First, DeepCoFFEA uses deep learning to train a pair of feature embedding networks that respectively map Tor and exit flows into a single low-dimensional space where correlated flows are similar; pairs of embedded flows can be compared at lower cost than pairs of full traces. Second, DeepCoFFEA uses amplification, dividing flows into short windows and using voting across these windows to significantly reduce false positives; the same embedding networks can be used with an increasing number of windows to independently lower the false positive rate. We conduct a comprehensive experimental analysis showing that DeepCoFFEA significantly outperforms state-of-the-art flow correlation attacks on Tor, e.g. 93% true positive rate versus at most 13% when tuned for high precision, with two orders of magnitude speedup over prior work. We also consider the effects of several potential countermeasures on DeepCoFFEA, finding that existing lightweight defenses are not sufficient to secure anonymity networks from this threat.

Abstract We introduce Generative Adversarial Networks for Data-Limited Fingerprinting (GANDaLF), a new deep-learning-based technique to perform Website Fingerprinting (WF) on Tor traffic. In contrast to most earlier work on deep-learning for WF, GANDaLF is intended to work with few training samples, and achieves this goal through the use of a Generative Adversarial Network to generate a large set of “fake” data that helps to train a deep neural network in distinguishing between classes of actual training data. We evaluate GANDaLF in low-data scenarios including as few as 10 training instances per site, and in multiple settings, including fingerprinting of website index pages and fingerprinting of non-index pages within a site. GANDaLF achieves closed-world accuracy of 87% with just 20 instances per site (and 100 sites) in standard WF settings. In particular, GANDaLF can outperform Var-CNN and Triplet Fingerprinting (TF) across all settings in subpage fingerprinting. For example, GANDaLF outperforms TF by a 29% margin and Var-CNN by 38% for training sets using 20 instances per site.

Website Fingerprinting (WF) is a type of traffic analysis attack that enables a local passive eavesdropper to infer the victim's activity, even when the traffic is protected by a VPN or an anonymity system like Tor. Leveraging a deep-learning classifier, a WF attacker can gain over 98% accuracy on Tor traffic. In this paper, we explore a novel defense, Mockingbird, based on the idea of adversarial examples that have been shown to undermine machine-learning classifiers in other domains. Since the attacker gets to design and train his attack classifier based on the defense, we first demonstrate that at a straightforward technique for generating adversarial-example based traces fails to protect against an attacker using adversarial training for robust classification. We then propose Mockingbird, a technique for generating traces that resists adversarial training by moving randomly in the space of viable traces and not following more predictable gradients. The technique drops the accuracy of the state-of-the-art attack hardened with adversarial training from 98% to 42-58% while incurring only 58% bandwidth overhead. The attack accuracy is generally lower than state-of-the-art defenses, and much lower when considering Top-2 accuracy, while incurring lower bandwidth overheads.

Abstract A passive local eavesdropper can leverage Website Fingerprinting (WF) to deanonymize the web browsing activity of Tor users. The value of timing information to WF has often been discounted in recent works due to the volatility of low-level timing information. In this paper, we more carefully examine the extent to which packet timing can be used to facilitate WF attacks. We first propose a new set of timing-related features based on burst-level characteristics to further identify more ways that timing patterns could be used by classifiers to identify sites. Then we evaluate the effectiveness of both raw timing and directional timing which is a combination of raw timing and direction in a deep-learning-based WF attack. Our closed-world evaluation shows that directional timing performs best in most of the settings we explored, achieving: (i) 98.4% in undefended Tor traffic; (ii) 93.5% on WTF-PAD traffic, several points higher than when only directional information is used; and (iii) 64.7% against onion sites, 12% higher than using only direction. Further evaluations in the open-world setting show small increases in both precision (+2%) and recall (+6%) with directional-timing on WTF-PAD traffic. To further investigate the value of timing information, we perform an information leakage analysis on our proposed handcrafted features. Our results show that while timing features leak less information than directional features, the information contained in each feature is mutually exclusive to one another and can thus improve the robustness of a classifier.

The website fingerprinting attack allows a low-resource attacker to compromise the privacy guarantees provided by privacy enhancing tools such as Tor. In response, researchers have proposed defenses aimed at confusing the classification tools used by attackers. As new, more powerful attacks are frequently developed, raw attack accuracy has proven inadequate as the sole metric used to evaluate these defenses. In response, two security metrics have been proposed that allow for evaluating defenses based on hand-crafted features often used in attacks. Recent state-of-the-art attacks, however, use deep learning models capable of automatically learning abstract feature representations, and thus the proposed metrics fall short once again. In this study we examine two security metrics and (1) show how these methods can be extended to evaluate deep learning-based website fingerprinting attacks, and (2) compare the security metrics and identify their shortcomings.

Over 8 million users rely on the Tor network each day to protect their anonymity online. Unfortunately, Tor has been shown to be vulnerable to the website fingerprinting attack, which allows an attacker to deduce the website a user is visiting based on patterns in their traffic. The state-of-the-art attacks leverage deep learning to achieve high classification accuracy using raw packet information. Work thus far, however, has examined only one type of media delivered over the Tor network: web pages, and mostly just home pages of sites. In this work, we instead investigate the fingerprintability of video content served over Tor. We collected a large new dataset of network traces for 50 YouTube videos of similar length. Our preliminary experiments utilizing a convolutional neural network model proposed in prior works has yielded promising classification results, achieving up to 55% accuracy. This shows the potential to unmask the individual videos that users are viewing over Tor, creating further privacy challenges to consider when defending against website fingerprinting attacks.

Website Fingerprinting (WF) attacks pose a serious threat to users' online privacy, including for users of the Tor anonymity system. By exploiting recent advances in deep learning, WF attacks like Deep Fingerprinting (DF) have reached up to 98% accuracy. The DF attack, however, requires large amounts of training data that needs to be updated regularly, making it less practical for the weaker attacker model typically assumed in WF. Moreover, research on WF attacks has been criticized for not demonstrating attack effectiveness under more realistic and more challenging scenarios. Most research on WF attacks assumes that the testing and training data have similar distributions and are collected from the same type of network at about the same time. In this paper, we examine how an attacker could leverage N-shot learning---a machine learning technique requiring just a few training samples to identify a given class---to reduce the effort of gathering and training with a large WF dataset as well as mitigate the adverse effects of dealing with different network conditions. In particular, we propose a new WF attack called Triplet Fingerprinting (TF) that uses triplet networks for N-shot learning. We evaluate this attack in challenging settings such as where the training and testing data are collected multiple years apart on different networks, and we find that the TF attack remains effective in such settings with 85% accuracy or better. We also show that the TF attack is also effective in the open world and outperforms traditional transfer learning. On top of that, the attack requires only five examples to recognize a website, making it dangerous in a wide variety of scenarios where gathering and training on a complete dataset would be impractical.

Website Fingerprinting (WF) is a traffic analysis attack that enables an eavesdropper to infer the victim's web activity even when encrypted and even when using the Tor anonymity system. Using deep learning classifiers, the attack can reach up to 98% accuracy. Existing WF defenses are either too expensive in terms of bandwidth and latency overheads (e.g. 2-3 times as large or slow) or ineffective against the latest attacks. In this work, we explore a novel defense based on the idea of adversarial examples that have been shown to undermine machine learning classifiers in other domains. Our Adversarial Traces defense adds padding to a Tor traffic trace in a manner that reliably fools the classifier into classifying it as coming from a different site. The technique drops the accuracy of the state-of-the-art attack from 98% to 60%, while incurring a reasonable 47% bandwidth overhead, showing its promise as a possible defense for Tor.