Training Robust Deep Neural Networks via Adversarial Noise Propagation

• Main Authors: Liu, Aishan, Liu, Xianglong, Zhang, Chongzhi, Yu, Hang, Liu, Qiang, Tao, Dacheng
• Format: Journal Article
• Language: English
• Published: 19.09.2019
• Subjects: Computer Science - Computer Vision and Pattern Recognition; Computer Science - Learning

In practice, deep neural networks have been found to be vulnerable to various types of noise, such as adversarial examples and corruption. Various adversarial defense methods have accordingly been developed to improve adversarial robustness for deep models. However, simply training on data mixed wit...