Deep learning adversarial attack defense method based on generative adversarial network

Abstract

The invention provides a deep learning adversarial attack defense method based on a generative adversarial network. The method comprises the following steps: step 1), based on the high performance ofthe generative adversarial network in learning sample distribution, designing a method for generating an adversarial example through the generative adversarial network, and after adding a target modelnetwork set TMi, enabling the sample generation based on a G network to become a multi-objective optimization problem; and the training for an AG-GAN model is mainly for the parameter training of thegenerative network G and a discrimination network D, and is divided into three modules; and step 2), using the adversarial example generated by the AG-GAN to train an attacked deep learning model, soas to improve the capability of the deep learning model of defending different types of adversarial examples. The deep learning adversarial attack defense method based on the generative adversarial network provided by the invention effectively improves the security.

Description

technical field [0001] The invention belongs to the security field of machine learning methods in the field of artificial intelligence. Aiming at the threat of adversarial sample attacks in deep learning methods in current machine learning, a deep learning adversarial attack defense method based on an adversarial generation network is proposed, which effectively improves its safety. Background technique [0002] Due to its good learning performance, the deep neural network model has been widely used in the real world, including computer vision, natural language processing, bioinformatics analysis, etc. Especially in the field of computer vision, driven by competitions such as ILSVRC and Kaggle, deep neural networks have shown more advanced performance than other machine learning methods. Graves et al. designed a new deep learning system, Differentiable Neural Computer (DNC), which introduces peripheral knowledge into deep learning. proposed a new type of neural network st...

AI Technical Summary

Problems solved by technology

[0005] In order to overcome the inability of existing technologies to solve deep learning-oriented adversarial sample attacks and low security, the present invention is based on an adversarial generative network-based deep learning adversarial attack defense method, which effectively improves its security

Images