Contrastive Neuron Pruning for Backdoor Defense

Yu Feng;Benteng Ma;Dongnan Liu;Yanning Zhang;Weidong Cai;Yong Xia
{"title":"Contrastive Neuron Pruning for Backdoor Defense","authors":"Yu Feng;Benteng Ma;Dongnan Liu;Yanning Zhang;Weidong Cai;Yong Xia","doi":"10.1109/TIP.2025.3539466","DOIUrl":null,"url":null,"abstract":"Recent studies have revealed that deep neural networks (DNNs) are susceptible to backdoor attacks, in which attackers insert a pre-defined backdoor into a DNN model by poisoning a few training samples. A small subset of neurons in DNN is responsible for activating this backdoor and pruning these backdoor-associated neurons has been shown to mitigate the impact of such attacks. Current neuron pruning techniques often face challenges in accurately identifying these critical neurons, and they typically depend on the availability of labeled clean data, which is not always feasible. To address these challenges, we propose a novel defense strategy called Contrastive Neuron Pruning (CNP). This approach is based on the observation that poisoned samples tend to cluster together and are distinguishable from benign samples in the feature space of a backdoored model. Given a backdoored model, we initially apply a reversed trigger to benign samples, generating multiple positive (benign-benign) and negative (benign-poisoned) feature pairs from the backdoored model. We then employ contrastive learning on these pairs to improve the separation between benign and poisoned features. Subsequently, we identify and prune neurons in the Batch Normalization layers that show significant response differences to the generated pairs. By removing these backdoor-associated neurons, CNP effectively defends against backdoor attacks while requiring the pruning of only about 1% of the total neurons. Comprehensive experiments conducted on various benchmarks validate the efficacy of CNP, demonstrating its robustness and effectiveness in mitigating backdoor attacks compared to existing methods.","PeriodicalId":94032,"journal":{"name":"IEEE transactions on image processing : a publication of the IEEE Signal Processing Society","volume":"34 ","pages":"1234-1245"},"PeriodicalIF":0.0000,"publicationDate":"2025-02-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"IEEE transactions on image processing : a publication of the IEEE Signal Processing Society","FirstCategoryId":"1085","ListUrlMain":"https://ieeexplore.ieee.org/document/10887056/","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0

Abstract

Recent studies have revealed that deep neural networks (DNNs) are susceptible to backdoor attacks, in which attackers insert a pre-defined backdoor into a DNN model by poisoning a few training samples. A small subset of neurons in DNN is responsible for activating this backdoor and pruning these backdoor-associated neurons has been shown to mitigate the impact of such attacks. Current neuron pruning techniques often face challenges in accurately identifying these critical neurons, and they typically depend on the availability of labeled clean data, which is not always feasible. To address these challenges, we propose a novel defense strategy called Contrastive Neuron Pruning (CNP). This approach is based on the observation that poisoned samples tend to cluster together and are distinguishable from benign samples in the feature space of a backdoored model. Given a backdoored model, we initially apply a reversed trigger to benign samples, generating multiple positive (benign-benign) and negative (benign-poisoned) feature pairs from the backdoored model. We then employ contrastive learning on these pairs to improve the separation between benign and poisoned features. Subsequently, we identify and prune neurons in the Batch Normalization layers that show significant response differences to the generated pairs. By removing these backdoor-associated neurons, CNP effectively defends against backdoor attacks while requiring the pruning of only about 1% of the total neurons. Comprehensive experiments conducted on various benchmarks validate the efficacy of CNP, demonstrating its robustness and effectiveness in mitigating backdoor attacks compared to existing methods.
求助全文
约1分钟内获得全文 求助全文
来源期刊
自引率
0.00%
发文量
0
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
确定
请完成安全验证×
copy
已复制链接
快去分享给好友吧!
我知道了
右上角分享
点击右上角分享
0
联系我们:info@booksci.cn Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。 Copyright © 2023 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术官方微信