Towards Imperceptible and Robust Adversarial Example Attacks against Neural Networks

Abstract

Machine learning systems based on deep neural networks, being able to produce state-of-the-art results on various perception tasks, have gained mainstream adoption in many applications. However, they are shown to be vulnerable to adversarial example attack, which generates malicious output by adding slight perturbations to the input. Previous adversarial example crafting methods, however, use simple metrics to evaluate the distances between the original examples and the adversarial ones, which could be easily detected by human eyes. In addition, these attacks are often not robust due to the inevitable noises and deviation in the physical world. In this work, we present a new adversarial example attack crafting method, which takes the human perceptual system into consideration and maximizes the noise tolerance of the crafted adversarial example. Experimental results demonstrate the efficacy of the proposed technique.

Adversarial example attacks, Robust and Imperceptible, Human perceptual system, Neural Networks

References in corpus (5)

• Explaining and Harnessing Adversarial Examples
• Intriguing properties of neural networks
• End to End Learning for Self-Driving Cars
• Robust Physical-World Attacks on Deep Learning Models
• Adversarial Perturbations Against Deep Neural Networks for Malware Classification

Cited by in corpus (7)

• Ten Years of Generative Adversarial Nets (GANs): A survey of the state-of-the-art
• Towards a Robust Deep Neural Network in Texts: A Survey
• There is Limited Correlation between Coverage and Robustness for Deep Neural Networks
• CapsAttacks: Robust and Imperceptible Adversarial Attacks on Capsule Networks
• Robustness of Rotation-Equivariant Networks to Adversarial Perturbations
• Delving into Data: Effectively Substitute Training for Black-box Attack
• Landmark Breaker: Obstructing DeepFake By Disturbing Landmark Extraction