从无标记数据中融合先验知识和后验知识,通过主动学习提取模型

Lijun Gao, Kai Liu, Wenjun Liu, Jiehong Wu, Xiao Jin
{"title":"从无标记数据中融合先验知识和后验知识,通过主动学习提取模型","authors":"Lijun Gao, Kai Liu, Wenjun Liu, Jiehong Wu, Xiao Jin","doi":"10.3233/jifs-239504","DOIUrl":null,"url":null,"abstract":"As machine learning models become increasingly integrated into practical applications and are made accessible via public APIs, the risk of model extraction attacks has gained prominence. This study presents an innovative and efficient approach to model extraction attacks, aimed at reducing query costs and enhancing attack effectiveness. The method begins by leveraging a pre-trained model to identify high-confidence samples from unlabeled datasets. It then employs unsupervised contrastive learning to thoroughly dissect the structural nuances of these samples, constructing a dataset of high quality that precisely mirrors a variety of features. A mixed information confidence strategy is employed to refine the query set, effectively probing the decision boundaries of the target model. By integrating consistency regularization and pseudo-labeling techniques, reliance on authentic labels is minimized, thus improving the feature extraction capabilities and predictive precision of the surrogate models. Evaluation on four major datasets reveals that the models crafted through this method bear a close functional resemblance to the original models, with a real-world API test success rate of 62.35%, which vouches for the method’s validity.","PeriodicalId":509313,"journal":{"name":"Journal of Intelligent & Fuzzy Systems","volume":null,"pages":null},"PeriodicalIF":0.0000,"publicationDate":"2024-03-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Model extraction via active learning by fusing prior and posterior knowledge from unlabeled data\",\"authors\":\"Lijun Gao, Kai Liu, Wenjun Liu, Jiehong Wu, Xiao Jin\",\"doi\":\"10.3233/jifs-239504\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"As machine learning models become increasingly integrated into practical applications and are made accessible via public APIs, the risk of model extraction attacks has gained prominence. This study presents an innovative and efficient approach to model extraction attacks, aimed at reducing query costs and enhancing attack effectiveness. The method begins by leveraging a pre-trained model to identify high-confidence samples from unlabeled datasets. It then employs unsupervised contrastive learning to thoroughly dissect the structural nuances of these samples, constructing a dataset of high quality that precisely mirrors a variety of features. A mixed information confidence strategy is employed to refine the query set, effectively probing the decision boundaries of the target model. By integrating consistency regularization and pseudo-labeling techniques, reliance on authentic labels is minimized, thus improving the feature extraction capabilities and predictive precision of the surrogate models. Evaluation on four major datasets reveals that the models crafted through this method bear a close functional resemblance to the original models, with a real-world API test success rate of 62.35%, which vouches for the method’s validity.\",\"PeriodicalId\":509313,\"journal\":{\"name\":\"Journal of Intelligent & Fuzzy Systems\",\"volume\":null,\"pages\":null},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2024-03-19\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Journal of Intelligent & Fuzzy Systems\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.3233/jifs-239504\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Journal of Intelligent & Fuzzy Systems","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.3233/jifs-239504","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0

摘要

随着机器学习模型越来越多地集成到实际应用中,并可通过公共应用程序接口访问,模型提取攻击的风险日益突出。本研究针对模型提取攻击提出了一种创新而高效的方法,旨在降低查询成本并提高攻击效果。该方法首先利用预先训练好的模型,从未标明的数据集中识别出高信度样本。然后,它采用无监督对比学习,彻底剖析这些样本结构上的细微差别,构建一个精确反映各种特征的高质量数据集。采用混合信息置信策略来完善查询集,从而有效探测目标模型的决策边界。通过整合一致性正则化和伪标签技术,最大限度地减少了对真实标签的依赖,从而提高了特征提取能力和代用模型的预测精度。在四个主要数据集上进行的评估表明,通过这种方法制作的模型在功能上与原始模型非常相似,实际 API 测试成功率高达 62.35%,这证明了该方法的有效性。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
查看原文
分享 分享
微信好友 朋友圈 QQ好友 复制链接
本刊更多论文
Model extraction via active learning by fusing prior and posterior knowledge from unlabeled data
As machine learning models become increasingly integrated into practical applications and are made accessible via public APIs, the risk of model extraction attacks has gained prominence. This study presents an innovative and efficient approach to model extraction attacks, aimed at reducing query costs and enhancing attack effectiveness. The method begins by leveraging a pre-trained model to identify high-confidence samples from unlabeled datasets. It then employs unsupervised contrastive learning to thoroughly dissect the structural nuances of these samples, constructing a dataset of high quality that precisely mirrors a variety of features. A mixed information confidence strategy is employed to refine the query set, effectively probing the decision boundaries of the target model. By integrating consistency regularization and pseudo-labeling techniques, reliance on authentic labels is minimized, thus improving the feature extraction capabilities and predictive precision of the surrogate models. Evaluation on four major datasets reveals that the models crafted through this method bear a close functional resemblance to the original models, with a real-world API test success rate of 62.35%, which vouches for the method’s validity.
求助全文
通过发布文献求助,成功后即可免费获取论文全文。 去求助
来源期刊
自引率
0.00%
发文量
0
期刊最新文献
Data-driven control of a five-bar parallel robot with compliant joints CycleGAN generated pneumonia chest x-ray images: Evaluation with vision transformer Robust image registration for analysis of multisource eye fundus images An efficient two-heuristic algorithm for the student-project allocation with preferences over projects Dynamic task scheduling in edge cloud systems using deep recurrent neural networks and environment learning approaches
×
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
现在去查看 取消
×
提示
确定
0
微信
客服QQ
Book学术公众号 扫码关注我们
反馈
×
意见反馈
请填写您的意见或建议
请填写您的手机或邮箱
已复制链接
已复制链接
快去分享给好友吧!
我知道了
×
扫码分享
扫码分享
Book学术官方微信
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术
文献互助 智能选刊 最新文献 互助须知 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 Book学术 All rights reserved.
ghs 京公网安备 11010802042870号 京ICP备2023020795号-1