Cyber-physical systems (CPS) have been increasingly attacked by hackers. Recent studies have shown that CPS are especially vulnerable to insider attacks, in which case the attacker has full knowledge of the systems configuration. To better prevent such types of attacks, we need to understand how insider attacks are generated. Typically, there are three critical aspects for a successful insider attack: (i) Maximize damage, (ii) Avoid detection and (iii) Minimize the attack cost. In this paper we propose a Stealthy Attack GEneration (SAGE) framework by formulizing a novel optimization problem considering these three objectives and the physical constraints of the CPS. By adding small worst-case perturbations to the system, the SAGE attack can generate significant damage, while remaining undetected by the systems monitoring algorithms. The proposed methodology is evaluated on several anomaly detection algorithms. The results show that SAGE attacks can cause severe damage while staying undetected and keeping the cost of an attack low. Our method can be accessed in the supplementary material of this paper to aid researcher and practitioners in the design and development of resilient CPS and detection algorithms.
相關內容
Machine Learning (ML) has proven to be effective in many application domains. However, ML methods can be vulnerable to adversarial attacks, in which an attacker tries to fool the classification/prediction mechanism by crafting the input data. In the case of ML-based Network Intrusion Detection Systems (NIDSs), the attacker might use their knowledge of the intrusion detection logic to generate malicious traffic that remains undetected. One way to solve this issue is to adopt adversarial training, in which the training set is augmented with adversarial traffic samples. This paper presents an adversarial training approach called GADoT, which leverages a Generative Adversarial Network (GAN) to generate adversarial DDoS samples for training. We show that a state-of-the-art NIDS with high accuracy on popular datasets can experience more than 60% undetected malicious flows under adversarial attacks. We then demonstrate how this score drops to 1.8% or less after adversarial training using GADoT.
Command, Control, Communication, and Intelligence (C3I) systems are increasingly used in critical civil and military domains for achieving information superiority, operational efficacy, and greater situational awareness. Unlike traditional systems facing widespread cyber-attacks, the sensitive nature of C3I tactical operations make their cybersecurity a critical concern. For instance, tampering or intercepting confidential information in military battlefields not only damages C3I operations, but also causes irreversible consequences such as loss of human lives and mission failures. Therefore, C3I systems have become a focal point for cyber adversaries. Moreover, technological advancements and modernization of C3I systems have significantly increased the potential risk of cyber-attacks on C3I systems. Consequently, cyber adversaries leverage highly sophisticated attack vectors to exploit security vulnerabilities in C3I systems. Despite the burgeoning significance of cybersecurity for C3I systems, the existing literature lacks a comprehensive review to systematize the body of knowledge on C3I systems' security. Therefore, in this paper, we have gathered, analyzed, and synthesized the state-of-the-art on the cybersecurity of C3I systems. In particular, this paper has identified security vulnerabilities, attack vectors, and countermeasures/defenses for C3I systems. Furthermore, our survey has enabled us to: (i) propose a taxonomy for security vulnerabilities, attack vectors and countermeasures; (ii) interrelate attack vectors with security vulnerabilities and countermeasures; and (iii) propose future research directions for advancing the state-of-the-art on the cybersecurity of C3I systems.
FPGAs are now used in public clouds to accelerate a wide range of applications, including many that operate on sensitive data such as financial and medical records. We present ShEF, a trusted execution environment (TEE) for cloud-based reconfigurable accelerators. ShEF is independent from CPU-based TEEs and allows secure execution under a threat model where the adversary can control all software running on the CPU connected to the FPGA, has physical access to the FPGA, and can compromise the FPGA interface logic of the cloud provider. ShEF provides a secure boot and remote attestation process that relies solely on existing FPGA mechanisms for root of trust. It also includes a Shield component that provides secure access to data while the accelerator is in use. The Shield is highly customizable and extensible, allowing users to craft a bespoke security solution that fits their accelerator's memory access patterns, bandwidth, and security requirements at minimum performance and area overheads. We describe a prototype implementation of ShEF for existing cloud FPGAs, map ShEF to a performant and secure storage application, and measure the performance benefits of customizable security using five additional accelerators.
Ever since Machine Learning as a Service (MLaaS) emerges as a viable business that utilizes deep learning models to generate lucrative revenue, Intellectual Property Right (IPR) has become a major concern because these deep learning models can easily be replicated, shared, and re-distributed by any unauthorized third parties. To the best of our knowledge, one of the prominent deep learning models - Generative Adversarial Networks (GANs) which has been widely used to create photorealistic image are totally unprotected despite the existence of pioneering IPR protection methodology for Convolutional Neural Networks (CNNs). This paper therefore presents a complete protection framework in both black-box and white-box settings to enforce IPR protection on GANs. Empirically, we show that the proposed method does not compromise the original GANs performance (i.e. image generation, image super-resolution, style transfer), and at the same time, it is able to withstand both removal and ambiguity attacks against embedded watermarks.
Adversarial attack is a technique for deceiving Machine Learning (ML) models, which provides a way to evaluate the adversarial robustness. In practice, attack algorithms are artificially selected and tuned by human experts to break a ML system. However, manual selection of attackers tends to be sub-optimal, leading to a mistakenly assessment of model security. In this paper, a new procedure called Composite Adversarial Attack (CAA) is proposed for automatically searching the best combination of attack algorithms and their hyper-parameters from a candidate pool of \textbf{32 base attackers}. We design a search space where attack policy is represented as an attacking sequence, i.e., the output of the previous attacker is used as the initialization input for successors. Multi-objective NSGA-II genetic algorithm is adopted for finding the strongest attack policy with minimum complexity. The experimental result shows CAA beats 10 top attackers on 11 diverse defenses with less elapsed time (\textbf{6 $\times$ faster than AutoAttack}), and achieves the new state-of-the-art on $l_{\infty}$, $l_{2}$ and unrestricted adversarial attacks.
As data are increasingly being stored in different silos and societies becoming more aware of data privacy issues, the traditional centralized training of artificial intelligence (AI) models is facing efficiency and privacy challenges. Recently, federated learning (FL) has emerged as an alternative solution and continue to thrive in this new reality. Existing FL protocol design has been shown to be vulnerable to adversaries within or outside of the system, compromising data privacy and system robustness. Besides training powerful global models, it is of paramount importance to design FL systems that have privacy guarantees and are resistant to different types of adversaries. In this paper, we conduct the first comprehensive survey on this topic. Through a concise introduction to the concept of FL, and a unique taxonomy covering: 1) threat models; 2) poisoning attacks and defenses against robustness; 3) inference attacks and defenses against privacy, we provide an accessible review of this important topic. We highlight the intuitions, key techniques as well as fundamental assumptions adopted by various attacks and defenses. Finally, we discuss promising future research directions towards robust and privacy-preserving federated learning.
There has been an ongoing cycle where stronger defenses against adversarial attacks are subsequently broken by a more advanced defense-aware attack. We present a new approach towards ending this cycle where we "deflect'' adversarial attacks by causing the attacker to produce an input that semantically resembles the attack's target class. To this end, we first propose a stronger defense based on Capsule Networks that combines three detection mechanisms to achieve state-of-the-art detection performance on both standard and defense-aware attacks. We then show that undetected attacks against our defense often perceptually resemble the adversarial target class by performing a human study where participants are asked to label images produced by the attack. These attack images can no longer be called "adversarial'' because our network classifies them the same way as humans do.
Deep neural networks (DNNs) are found to be vulnerable against adversarial examples, which are carefully crafted inputs with a small magnitude of perturbation aiming to induce arbitrarily incorrect predictions. Recent studies show that adversarial examples can pose a threat to real-world security-critical applications: a "physical adversarial Stop Sign" can be synthesized such that the autonomous driving cars will misrecognize it as others (e.g., a speed limit sign). However, these image-space adversarial examples cannot easily alter 3D scans of widely equipped LiDAR or radar on autonomous vehicles. In this paper, we reveal the potential vulnerabilities of LiDAR-based autonomous driving detection systems, by proposing an optimization based approach LiDAR-Adv to generate adversarial objects that can evade the LiDAR-based detection system under various conditions. We first show the vulnerabilities using a blackbox evolution-based algorithm, and then explore how much a strong adversary can do, using our gradient-based approach LiDAR-Adv. We test the generated adversarial objects on the Baidu Apollo autonomous driving platform and show that such physical systems are indeed vulnerable to the proposed attacks. We also 3D-print our adversarial objects and perform physical experiments to illustrate that such vulnerability exists in the real world. Please find more visualizations and results on the anonymous website: //sites.google.com/view/lidar-adv.
There is a rising interest in studying the robustness of deep neural network classifiers against adversaries, with both advanced attack and defence techniques being actively developed. However, most recent work focuses on discriminative classifiers, which only model the conditional distribution of the labels given the inputs. In this paper we propose the deep Bayes classifier, which improves classical naive Bayes with conditional deep generative models. We further develop detection methods for adversarial examples, which reject inputs that have negative log-likelihood under the generative model exceeding a threshold pre-specified using training data. Experimental results suggest that deep Bayes classifiers are more robust than deep discriminative classifiers, and the proposed detection methods achieve high detection rates against many recently proposed attacks.
Reinforcement learning (RL) has advanced greatly in the past few years with the employment of effective deep neural networks (DNNs) on the policy networks. With the great effectiveness came serious vulnerability issues with DNNs that small adversarial perturbations on the input can change the output of the network. Several works have pointed out that learned agents with a DNN policy network can be manipulated against achieving the original task through a sequence of small perturbations on the input states. In this paper, we demonstrate furthermore that it is also possible to impose an arbitrary adversarial reward on the victim policy network through a sequence of attacks. Our method involves the latest adversarial attack technique, Adversarial Transformer Network (ATN), that learns to generate the attack and is easy to integrate into the policy network. As a result of our attack, the victim agent is misguided to optimise for the adversarial reward over time. Our results expose serious security threats for RL applications in safety-critical systems including drones, medical analysis, and self-driving cars.
注冊地址： 北京市海淀區羊坊店路18號2幢3層301-191