In this paper, we propose a planning framework to generate a defense strategy against an attacker who is working in an environment where a defender can operate without the attacker's knowledge. The objective of the defender is to covertly guide the attacker to a trap state from which the attacker cannot achieve their goal. Further, the defender is constrained to achieve its goal within K number of steps, where K is calculated as a pessimistic lower bound within which the attacker is unlikely to suspect a threat in the environment. Such a defense strategy is highly useful in real world systems like honeypots or honeynets, where an unsuspecting attacker interacts with a simulated production system while assuming it is the actual production system. Typically, the interaction between an attacker and a defender is captured using game theoretic frameworks. Our problem formulation allows us to capture it as a much simpler infinite horizon discounted MDP, in which the optimal policy for the MDP gives the defender's strategy against the actions of the attacker. Through empirical evaluation, we show the merits of our problem formulation.
相關內容
Recent advances in deep generative models have led to the development of methods capable of synthesizing high-quality, realistic images. These models pose threats to society due to their potential misuse. Prior research attempted to mitigate these threats by detecting generated images, but the varying traces left by different generative models make it challenging to create a universal detector capable of generalizing to new, unseen generative models. In this paper, we propose to inject a universal adversarial signature into an arbitrary pre-trained generative model, in order to make its generated contents more detectable and traceable. First, the imperceptible optimal signature for each image can be found by a signature injector through adversarial training. Subsequently, the signature can be incorporated into an arbitrary generator by fine-tuning it with the images processed by the signature injector. In this way, the detector corresponding to the signature can be reused for any fine-tuned generator for tracking the generator identity. The proposed method is validated on the FFHQ and ImageNet datasets with various state-of-the-art generative models, consistently showing a promising detection rate. Code will be made publicly available at \url{//github.com/zengxianyu/genwm}.
This research article critically examines the potential risks and implications arising from the malicious utilization of large language models(LLM), focusing specifically on ChatGPT and Google's Bard. Although these large language models have numerous beneficial applications, the misuse of this technology by cybercriminals for creating offensive payloads and tools is a significant concern. In this study, we systematically generated implementable code for the top-10 MITRE Techniques prevalent in 2022, utilizing ChatGPT, and conduct a comparative analysis of its performance with Google's Bard. Our experimentation reveals that ChatGPT has the potential to enable attackers to accelerate the operation of more targeted and sophisticated attacks. Additionally, the technology provides amateur attackers with more capabilities to perform a wide range of attacks and empowers script kiddies to develop customized tools that contribute to the acceleration of cybercrime. Furthermore, LLMs significantly benefits malware authors, particularly ransomware gangs, in generating sophisticated variants of wiper and ransomware attacks with ease. On a positive note, our study also highlights how offensive security researchers and pentesters can make use of LLMs to simulate realistic attack scenarios, identify potential vulnerabilities, and better protect organizations. Overall, we conclude by emphasizing the need for increased vigilance in mitigating the risks associated with LLMs. This includes implementing robust security measures, increasing awareness and education around the potential risks of this technology, and collaborating with security experts to stay ahead of emerging threats.
Despite their impressive performance in classification, neural networks are known to be vulnerable to adversarial attacks. These attacks are small perturbations of the input data designed to fool the model. Naturally, a question arises regarding the potential connection between the architecture, settings, or properties of the model and the nature of the attack. In this work, we aim to shed light on this problem by focusing on the implicit bias of the neural network, which refers to its inherent inclination to favor specific patterns or outcomes. Specifically, we investigate one aspect of the implicit bias, which involves the essential Fourier frequencies required for accurate image classification. We conduct tests to assess the statistical relationship between these frequencies and those necessary for a successful attack. To delve into this relationship, we propose a new method that can uncover non-linear correlations between sets of coordinates, which, in our case, are the aforementioned frequencies. By exploiting the entanglement between intrinsic dimension and correlation, we provide empirical evidence that the network bias in Fourier space and the target frequencies of adversarial attacks are closely tied.
With the emergence of more powerful large language models (LLMs), such as ChatGPT and GPT-4, in-context learning (ICL) has gained significant prominence in leveraging these models for specific tasks by utilizing data-label pairs as precondition prompts. While incorporating demonstrations can greatly enhance the performance of LLMs across various tasks, it may introduce a new security concern: attackers can manipulate only the demonstrations without changing the input to perform an attack. In this paper, we investigate the security concern of ICL from an adversarial perspective, focusing on the impact of demonstrations. We propose an ICL attack based on TextAttack, which aims to only manipulate the demonstration without changing the input to mislead the models. Our results demonstrate that as the number of demonstrations increases, the robustness of in-context learning would decreases. Furthermore, we also observe that adversarially attacked demonstrations exhibit transferability to diverse input examples. These findings emphasize the critical security risks associated with ICL and underscore the necessity for extensive research on the robustness of ICL, particularly given its increasing significance in the advancement of LLMs.
The release of ChatGPT, Bard, and other large language model (LLM)-based chatbots has drawn huge attention on foundations models worldwide. There is a growing trend that foundation models will serve as the fundamental building blocks for most of the future AI systems. However, incorporating foundation models in AI systems raises significant concerns about responsible AI due to their black box nature and rapidly advancing super-intelligence. Additionally, the foundation model's growing capabilities can eventually absorb the other components of AI systems, introducing the moving boundary and interface evolution challenges in architecture design. To address these challenges, this paper proposes a pattern-oriented responsible-AI-by-design reference architecture for designing foundation model-based AI systems. Specially, the paper first presents an architecture evolution of AI systems in the era of foundation models, from "foundation-model-as-a-connector" to "foundation-model-as-a-monolithic architecture". The paper then identifies the key design decision points and proposes a pattern-oriented reference architecture to provide reusable responsible-AI-by-design architectural solutions to address the new architecture evolution and responsible AI challenges. The patterns can be embedded as product features of foundation model-based AI systems and can enable organisations to capitalise on the potential of foundation models while minimising associated risks.
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.
While existing work in robust deep learning has focused on small pixel-level $\ell_p$ norm-based perturbations, this may not account for perturbations encountered in several real world settings. In many such cases although test data might not be available, broad specifications about the types of perturbations (such as an unknown degree of rotation) may be known. We consider a setup where robustness is expected over an unseen test domain that is not i.i.d. but deviates from the training domain. While this deviation may not be exactly known, its broad characterization is specified a priori, in terms of attributes. We propose an adversarial training approach which learns to generate new samples so as to maximize exposure of the classifier to the attributes-space, without having access to the data from the test domain. Our adversarial training solves a min-max optimization problem, with the inner maximization generating adversarial perturbations, and the outer minimization finding model parameters by optimizing the loss on adversarial perturbations generated from the inner maximization. We demonstrate the applicability of our approach on three types of naturally occurring perturbations -- object-related shifts, geometric transformations, and common image corruptions. Our approach enables deep neural networks to be robust against a wide range of naturally occurring perturbations. We demonstrate the usefulness of the proposed approach by showing the robustness gains of deep neural networks trained using our adversarial training on MNIST, CIFAR-10, and a new variant of the CLEVR dataset.
There is a recent large and growing interest in generative adversarial networks (GANs), which offer powerful features for generative modeling, density estimation, and energy function learning. GANs are difficult to train and evaluate but are capable of creating amazingly realistic, though synthetic, image data. Ideas stemming from GANs such as adversarial losses are creating research opportunities for other challenges such as domain adaptation. In this paper, we look at the field of GANs with emphasis on these areas of emerging research. To provide background for adversarial techniques, we survey the field of GANs, looking at the original formulation, training variants, evaluation methods, and extensions. Then we survey recent work on transfer learning, focusing on comparing different adversarial domain adaptation methods. Finally, we take a look forward to identify open research directions for GANs and domain adaptation, including some promising applications such as sensor-based human behavior modeling.
Most existing works in visual question answering (VQA) are dedicated to improving the accuracy of predicted answers, while disregarding the explanations. We argue that the explanation for an answer is of the same or even more importance compared with the answer itself, since it makes the question and answering process more understandable and traceable. To this end, we propose a new task of VQA-E (VQA with Explanation), where the computational models are required to generate an explanation with the predicted answer. We first construct a new dataset, and then frame the VQA-E problem in a multi-task learning architecture. Our VQA-E dataset is automatically derived from the VQA v2 dataset by intelligently exploiting the available captions. We have conducted a user study to validate the quality of explanations synthesized by our method. We quantitatively show that the additional supervision from explanations can not only produce insightful textual sentences to justify the answers, but also improve the performance of answer prediction. Our model outperforms the state-of-the-art methods by a clear margin on the VQA v2 dataset.
注冊地址： 北京市海淀區羊坊店路18號2幢3層301-191