arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2205.09167cs.CRcs.AI
收藏

基于反向分布的贝叶斯神经网络后门攻击

Backdoor Attacks on Bayesian Neural Networks using Reverse Distribution

University of Florida(佛罗里达大学)

Zhixin Pan, Prabhat Mishra

摘要:

由于成本和上市时间的限制,许多行业将机器学习模型(ML)的训练过程外包给第三方云服务提供商,通常称为ML-asa-Service(MLaaS)。MLaaS为对手创造了机会,为用户提供后门ML模型,仅在极其罕见的情况下(攻击者选择的)产生不正确的预测。贝叶斯神经网络(BNN)具有固有的免疫力,对后门攻击,因为权重被设计为边际分布,以量化的不确定性。在本文中,我们提出了一种新的后门攻击的基础上有效的学习和有针对性地利用反向分布。本文作出了三个重要贡献。(1)据我们所知,这是第一个后门攻击,可以有效地打破BNN的鲁棒性。(2)当触发器被激活时,我们产生反向分布来抵消原始分布。(3)我们提出了一个有效的解决方案,合并BNN的概率分布。在不同基准数据集上的实验结果表明,我们提出的攻击可以达到100%的攻击成功率(ASR),而最先进的攻击的ASR低于60%。

英文摘要:

Due to cost and time-to-market constraints, many industries outsource the training process of machine learning models (ML) to third-party cloud service providers, popularly known as ML-asa-Service (MLaaS). MLaaS creates opportunity for an adversary to provide users with backdoored ML models to produce incorrect predictions only in extremely rare (attacker-chosen) scenarios. Bayesian neural networks (BNN) are inherently immune against backdoor attacks since the weights are designed to be marginal distributions to quantify the uncertainty. In this paper, we propose a novel backdoor attack based on effective learning and targeted utilization of reverse distribution. This paper makes three important contributions. (1) To the best of our knowledge, this is the first backdoor attack that can effectively break the robustness of BNNs. (2) We produce reverse distributions to cancel the original distributions when the trigger is activated. (3) We propose an efficient solution for merging probability distributions in BNNs. Experimental results on diverse benchmark datasets demonstrate that our proposed attack can achieve the attack success rate (ASR) of 100%, while the ASR of the state-of-the-art attacks is lower than 60%.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑