Lightweight DCT-Like Domain Forensics Model for Adversarial Example

2021 ◽  
pp. 265-279
Author(s):  
Junjie Zhao ◽  
Jinwei Wang
Keyword(s):  
Adversarial Example

HRAE: Hardware-assisted Randomization against Adversarial Example Attacks

2020 ◽  
Author(s):  
Jiliang Zhang ◽  
Shuang Peng ◽  
Yupeng Hu ◽  
Fei Peng ◽  
Wei Hu ◽  
...  
Keyword(s):  
Adversarial Example

scholarly journals Adversarial example detection for DNN models: a review and experimental comparison

2022 ◽  
Author(s):  
Ahmed Aldahdooh ◽  
Wassim Hamidouche ◽  
Sid Ahmed Fezza ◽  
Olivier Déforges
Keyword(s):  
Experimental Comparison ◽  
Adversarial Example

Joint Adversarial Example and False Data Injection Attacks for State Estimation in Power Systems

2021 ◽  
pp. 1-15
Author(s):  
Jiwei Tian ◽  
Buhong Wang ◽  
Zhen Wang ◽  
Kunrui Cao ◽  
Jing Li ◽  
...  
Keyword(s):  
Power Systems ◽  
State Estimation ◽  
False Data Injection ◽  
Injection Attacks ◽  
Adversarial Example

scholarly journals Two Improved Methods of Generating Adversarial Examples against Faster R-CNNs for Tram Environment Perception Systems

Complexity ◽  
2020 ◽  
Vol 2020 ◽  
pp. 1-10
Author(s):  
Shize Huang ◽  
Xiaowen Liu ◽  
Xiaolu Yang ◽  
Zhaoxin Zhang ◽  
Lingyu Yang
Keyword(s):  
Neural Networks ◽  
Deep Learning ◽  
Gradient Descent ◽  
Learning Networks ◽  
Adversarial Examples ◽  
Environment Perception ◽  
Projected Gradient Descent ◽  
Adversarial Example ◽  
Improved Methods ◽  
Growing Neural Networks

Trams have increasingly deployed object detectors to perceive running conditions, and deep learning networks have been widely adopted by those detectors. Growing neural networks have incurred severe attacks such as adversarial example attacks, imposing threats to tram safety. Only if adversarial attacks are studied thoroughly, researchers can come up with better defence methods against them. However, most existing methods of generating adversarial examples have been devoted to classification, and none of them target tram environment perception systems. In this paper, we propose an improved projected gradient descent (PGD) algorithm and an improved Carlini and Wagner (C&W) algorithm to generate adversarial examples against Faster R-CNN object detectors. Experiments verify that both algorithms can successfully conduct nontargeted and targeted white-box digital attacks when trams are running. We also compare the performance of the two methods, including attack effects, similarity to clean images, and the generating time. The results show that both algorithms can generate adversarial examples within 220 seconds, a much shorter time, without decrease of the success rate.

Integration of statistical detector and Gaussian noise injection detector for adversarial example detection in deep neural networks

2019 ◽  
Vol 78 (14) ◽  
pp. 20409-20429 ◽  
Author(s):  
Weiqi Fan ◽  
Guangling Sun ◽  
Yuying Su ◽  
Zhi Liu ◽  
Xiaofeng Lu
Keyword(s):  
Neural Networks ◽  
Gaussian Noise ◽  
Deep Neural Networks ◽  
Noise Injection ◽  
Adversarial Example
Sign in / Sign up

Export Citation Format

Share Document