A Black-Box Approach to Generate Adversarial Examples Against Deep Neural Networks for High Dimensional Input

Song, Chengru; Xu, Changqiao; Yang, Shujie; Zhou, Zan; Gong, Changhui

doi:10.1109/dsc.2019.00078

Cited by 3 publications

(2 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Szegedy et al [12] first pointed out a major weakness of DNNs in the context of image classification: by adding adversarial examples (e.g., small perturbations that the human eyes may not be able to perceive) to the input samples, the neural network classifier may be fooled, yielding inaccurate predictions on input image samples. Moreover, such a perturbation may propagate between different DNN models [13], increasing the probability of being fooled. These misclassified samples are called adversarial samples.…”

Section: B Related Workmentioning

confidence: 99%

Attacking Spectrum Sensing With Adversarial Deep Learning in Cognitive Radio-Enabled Internet of Things

et al. 2023

View full text Add to dashboard Cite

Cognitive radio-based Internet of Things (CR-IoT) network provides a solution for IoT devices to efficiently utilize spectrum resources. Spectrum sensing is a critical problem in CR-IoT network, which has been investigated extensively based on deep learning (DL). Despite the unique advantages of DL in spectrum sensing, the black-box and unexplained properties of deep neural networks may lead to many security risks. This article considers the fusion of traditional interference methods and data poisoning which is an attack method on the training data of a machine learning tool. We propose a new adversarial attack for reducing the sensing accuracy in DL-based spectrum sensing systems. We introduce a novel design of jamming waveform whose interference capability is reinforced by data poisoning. Simulation results show that significant performance enhancement and higher mobility can be achieved compared with traditional white-box attack methods.

show abstract

Section: B Related Workmentioning

confidence: 99%

Attacking Spectrum Sensing With Adversarial Deep Learning in Cognitive Radio-Enabled Internet of Things

et al. 2023

View full text Add to dashboard Cite

show abstract

“…adversarial examples are then constructed using the known substitute parameters. As a boundary-based attack, the proposed approach bySong et al (2019) depends on computing the boundary of a classification result by utilizing a combination of linear fine-grained and Fibonacci searches. a zeroth-order algorithm byChen et al (2017) is utilized to estimate the gradient.…”

mentioning

confidence: 99%

Enhancing Trustworthy Deep Learning for Image Classification against Evasion Attacks: A systematic literature review

Akhtom,

Singh,

Ying

2023

Preprint

View full text Add to dashboard Cite

Deep learning (DL) algorithms have demonstrated remarkable performance in image classification , making them essential for critical applications. However, recent research has highlighted the vulnerability of DL to cyber-attacks, such as data poisoning and evasion attacks, which can significantly degrade its performance. Addressing this issue and developing trustworthy and robust DL models against such attacks remain ongoing research challenges. This study aims to evaluate the state-of-the-art Adversarial Machine Learning techniques (AML) in image classification through a systematic literature review encompassing 49 different studies conducted over the last decade. By conducting this review, potential obstacles in building robust DL models capable of withstanding evasion attacks are identified. The comprehensive analysis of existing literature is crucial to explore the impact of adversarial attacks on the security and resilience of DL models , which aligns with the regulations set forth by the European Commission’s AI High-Level Expert Group (HLEG) to ensure the reliability of artificial intelligence. Furthermore, this review examines the key features of adversarial attacks and proposes a reliable and trustworthy framework based on AML. This framework aims to effectively mitigate previously unknown attacks, resulting in reliable and explainable predictions.

show abstract

Enhancing trustworthy deep learning for image classification against evasion attacks: a systematic literature review

Akhtom,

Singh,

XinYing

2024

Artif Intell Rev

View full text Add to dashboard Cite

In the rapidly evolving field of Deep Learning (DL), the trustworthiness of models is essential for their effective application in critical domains like healthcare and autonomous systems. Trustworthiness in DL encompasses aspects such as reliability, fairness, and transparency, which are crucial for its real-world impact and acceptance. However, the development of trustworthy DL models faces significant challenges. This is notably due to adversarial examples, a sophisticated form of evasion attack in adversarial machine learning (AML), which subtly alter inputs to deceive these models and pose a major threat to their safety and reliability. The current body of research primarily focuses on defensive measures, such as enhancing the robustness of models or implementing explainable AI techniques. However, this approach often neglects to address the fundamental vulnerabilities that adversaries exploit. As a result, the field tends to concentrate more on counteracting measures rather than gaining an in-depth understanding of the vulnerabilities and attack strategies inherent in DL systems. This gap in comprehensive understanding impedes the formulation of effective defense mechanisms. This research aims to shift the focus from predominantly defensive strategies toward a more extensive comprehension of adversarial techniques and the innate vulnerabilities of DL models. We undertake this by conducting a thorough systematic literature review, encompassing 49 diverse studies from the previous decade. Our findings reveal the key characteristics of adversarial examples that enable their success against image classification-based DL models. Building on these insights, we propose the Transferable Pretrained Adversarial Deep Learning framework (TPre-ADL). This conceptual model aims to rectify the deficiencies in current defense strategies by incorporating the analyzed traits of adversarial examples, potentially enhancing the robustness and trustworthiness of DL models.

show abstract

A Black-Box Approach to Generate Adversarial Examples Against Deep Neural Networks for High Dimensional Input

Cited by 3 publications

References 9 publications

Attacking Spectrum Sensing With Adversarial Deep Learning in Cognitive Radio-Enabled Internet of Things

Attacking Spectrum Sensing With Adversarial Deep Learning in Cognitive Radio-Enabled Internet of Things

Enhancing Trustworthy Deep Learning for Image Classification against Evasion Attacks: A systematic literature review

Enhancing trustworthy deep learning for image classification against evasion attacks: a systematic literature review

Contact Info

Product

Resources

About