Generating Adversarial Samples with Convolutional Neural Network

Abstract

Deep learning has become a hot research direction in the field of computer vision, and has been widely applied in the fields of intelligent transportation, intelligent security and so on. Because deep learning is vulnerable to adversarial samples, therefore poses a great threat to some safety-sensitive applications such as autonomous driving. In order to study the application of convolutional neural networks in adversarial sample generation and to lay the foundation for future research adversarial sample characteristics, we propose a convolutional neural network for generating adversarial samples, which can successfully fool the deep learning model.