A defense method against backdoor attacks on neural networks

作者：

Highlights：

• Background on scale-free structures as complex networks and applications to ANNs.

• Previous works on detecting and mitigating techniques against different types of attacks to ANNs.

• A new method based on scale-free networks to make the feed-forward neural network robust.

• Experiments has shown the effectiveness of the proposed method against backdoor attacks.

摘要

•Background on scale-free structures as complex networks and applications to ANNs.•Previous works on detecting and mitigating techniques against different types of attacks to ANNs.•A new method based on scale-free networks to make the feed-forward neural network robust.•Experiments has shown the effectiveness of the proposed method against backdoor attacks.

论文关键词：Feed-forward neural networks,Backdoor attacks,Scale-free networks

论文评审过程：Received 18 November 2021, Revised 24 September 2022, Accepted 4 October 2022, Available online 13 October 2022, Version of Record 20 October 2022.

论文官网地址：https://doi.org/10.1016/j.eswa.2022.118990