Counteracting adversarial attacks in autonomous driving

Sun, Qi; Rao, Arjun Ashok; Yao, Xufeng; Yu, Bei

doi:10.1145/3400302.3415758

Cited by 15 publications

(4 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The proposed defense strategy was based on adversarial training with a novel regularization term, which considered the local smoothness and stereo information. The evaluation results showed that their approach was more effective than a regular adversarial training approach and it improved the detection performance of the original model as well [57].…”

Section: Defense Methods Introduced On Autonomous Vehicles Against Ad...mentioning

confidence: 99%

Adversarial Attacks and Defense Technologies on Autonomous Vehicles: A Review

Mahima

Ayoob

Poravi

2021

Applied Computer Systems

View full text Add to dashboard Cite

In recent years, various domains have been influenced by the rapid growth of machine learning. Autonomous driving is an area that has tremendously developed in parallel with the advancement of machine learning. In autonomous vehicles, various machine learning components are used such as traffic lights recognition, traffic sign recognition, limiting speed and pathfinding. For most of these components, computer vision technologies with deep learning such as object detection, semantic segmentation and image classification are used. However, these machine learning models are vulnerable to targeted tensor perturbations called adversarial attacks, which limit the performance of the applications. Therefore, implementing defense models against adversarial attacks has become an increasingly critical research area. The paper aims at summarising the latest adversarial attacks and defense models introduced in the field of autonomous driving with machine learning technologies up until mid-2021.

show abstract

Section: Defense Methods Introduced On Autonomous Vehicles Against Ad...mentioning

confidence: 99%

Adversarial Attacks and Defense Technologies on Autonomous Vehicles: A Review

Mahima

Ayoob

Poravi

2021

Applied Computer Systems

View full text Add to dashboard Cite

show abstract

“…These concerns have prompted adversarial attacks, revealing model vulnerabilities. The attacks such as FGSM attack and PGD attack tend to mislead the model to misclassify the object orientation of the vehicle or even could incorrectly perceive class 'Car' as class 'Background' potentially leading to an accident [29] .…”

Section: B Effects Of Adversarial Attacksmentioning

confidence: 99%

Threats of Adversarial Attacks on Deep Learning

Regmi,

Regmi

2023

IJRASET

View full text Add to dashboard Cite

AI, accompanied with Deep Learning, is a growing technological innovation which has an extensible reach to various applications. The complex computational models i.e. Deep Neural Networks (DNNs) excel in natural language processing, image recognition, computer vision, and autonomous systems. Despite the merits, these intricate networks are vulnerable to adversarial attacks (such as black-box and white-box) which can be the primary threats to the AI-DL integration. For their mitigation, adversarial training, defense distillations and other defenses are introduced. This paper deals with the impediments posed by adversarial attacks on Deep Learning, and lists out the possible solutions to reduce their impacts.

show abstract

“…Such malicious examples can mislead Deep Learning (DL) models into making wrong predictions without being perceived by human beings. Due to the critical threat, adversarial examples have been becoming a great challenge when applying deep learning in security-sensitive scenarios [2]- [7] such as autonomous driving [7]. Compared with white-box adversarial example generation attacks [1], [5], [8]- [11] that allow adversaries to access the architecture and parameters of the target model, blackbox attacks [12]- [20] are more threatening and practical in real applications, where an adversary can only query the target model via application programming interfaces (APIs).…”

Section: Introductionmentioning

confidence: 99%

Local Black-box Adversarial Attacks: A Query Efficient Approach

Xiang¹,

Liu²,

Guo³

et al. 2021

Preprint

View full text Add to dashboard Cite

Adversarial attacks have threatened the application of deep neural networks in security-sensitive scenarios. Most existing blackbox attacks fool the target model by interacting with it many times and producing global perturbations. However, global perturbations change the smooth and insignificant background, which not only makes the perturbation more easily be perceived but also increases the query overhead. In this paper, we propose a novel framework to perturb the discriminative areas of clean examples only within limited queries in black-box attacks. Our framework is constructed based on two types of transferability. The first one is the transferability of model interpretations. Based on this property, we identify the discriminative areas of a given clean example easily for local perturbations. The second is the transferability of adversarial examples. It helps us to produce a local pre-perturbation for improving query efficiency. After identifying the discriminative areas and pre-perturbing, we generate the final adversarial examples from the pre-perturbed example by querying the targeted model with two kinds of black-box attack techniques, i.e., gradient estimation and random search. We conduct extensive experiments to show that our framework can significantly improve the query efficiency during black-box perturbing with a high attack success rate. Experimental results show that our attacks outperform state-of-the-art black-box attacks under various system settings.

show abstract

Counteracting adversarial attacks in autonomous driving

Cited by 15 publications

References 19 publications

Adversarial Attacks and Defense Technologies on Autonomous Vehicles: A Review

Adversarial Attacks and Defense Technologies on Autonomous Vehicles: A Review

Threats of Adversarial Attacks on Deep Learning

Local Black-box Adversarial Attacks: A Query Efficient Approach

Contact Info

Product

Resources

About