Brain-inspired reverse adversarial examples

Shaokai Ye Tsinghua Univerity, China Sia Huat Tan Tsinghua Univerity, China Kaidi Xu Northeastern University, USA Yanzhi Wang Northeastern University, USA Chenglong Bao Tsinghua Univerity, China Kaisheng Ma Tsinghua Univerity, China

Machine Learning mathscidoc:2206.41011

A human does not have to see all elephants to recognize an animal as an elephant. On contrast, current state-of-the-art deep learning approaches heavily depend on the variety of training samples and the capacity of the network. In practice, the size of network is always limited and it is impossible to access all the data samples. Under this circumstance, deep learning models are extremely fragile to human-imperceivable adversarial examples, which impose threats to all safety critical systems. Inspired by the association and attention mechanisms of the human brain, we propose reverse adversarial examples method that can greatly improve models' robustness on unseen data. Experiments show that our reverse adversarial method can improve accuracy on average 19.02% on ResNet18, MobileNet, and VGG16 on unseen data transformation. Besides, the proposed method is also applicable to compressed models and shows potential to compensate the robustness drop brought by model quantization - an absolute 30.78% accuracy improvement.
No keywords uploaded!
[ Download ] [ 2022-06-16 15:57:22 uploaded by Baocl ] [ 36 downloads ] [ 0 comments ]
  title={Brain-inspired reverse adversarial examples},
  author={Shaokai Ye, Sia Huat Tan, Kaidi Xu, Yanzhi Wang, Chenglong Bao, and Kaisheng Ma},
Shaokai Ye, Sia Huat Tan, Kaidi Xu, Yanzhi Wang, Chenglong Bao, and Kaisheng Ma. Brain-inspired reverse adversarial examples. 2019.
Please log in for comment!
Contact us: | Copyright Reserved