Chernoff信息作为对抗分类和成员优势的隐私约束
Chernoff Information as a Privacy Constraint for Adversarial Classification and Membership Advantage
AI总结:
本文研究了基于Chernoff信息的隐私度量,旨在刻画最优分类器性能。对抗分类关注在二分类中最小化错误概率,而经典假设检验将虚警和误检概率分开处理,导致非对称最优误差指数。本文则刻画了ε-差分隐私(DP)、一个错误概率条件下的最优误差指数以及最优平均误差指数之间的关系,通过Radon-Nikodym导数重新推导Chernoff DP,并将其与KL-DP关系证明Chernoff DP介于两者之间。最后,我们通过数值评估展示Chernoff信息在隐私参数函数下优于KL散度,特别是在Laplace机制下对抗攻击的影响。
摘要:
本文研究了一种基于Chernoff信息的隐私度量,其重要性在于刻画最优分类器性能。对抗分类关注在二分类中最小化错误概率。经典假设检验将虚警和误检概率分开处理,导致非对称最优误差指数。本文则刻画了ε-差分隐私(DP)、一个错误概率条件下的最优误差指数以及最优平均误差指数之间的关系。通过Radon-Nikodym导数重新推导Chernoff DP,并将其与KL-DP关系证明Chernoff DP介于两者之间。我们随后展示了数值评估结果,证明Chernoff信息在隐私参数ε和Laplace机制下的对抗攻击影响下优于KL散度。最后,我们引入了基于Chernoff DP的新上界,用于成员推理攻击中的对手优势,并与文献中基于(ε,δ)-DP的现有替代方法进行了数值比较。
英文摘要:
This work investigates a privacy metric based on Chernoff information motivated by its importance in characterizing the optimal classifier's performance. Adversarial classification centers on minimizing the probability of error when deciding between two classes in the binary setting. Classical hypothesis testing treats false alarm and mis-detection probabilities separately, resulting in asymmetric optimal error exponents. Here, we instead characterize the relationship between $\varepsilon-$differential privacy (DP), the optimal error exponent of one error probability conditioned on the other, and the optimal average error exponent. Thus, we re-derive Chernoff DP in connection with $\varepsilon-$DP using the Radon-Nikodym derivative and establish its relationship with Kullback-Leibler (KL) DP to prove that Chernoff DP is sandwiched between the two. We then present numerical evaluations demonstrating that Chernoff information outperforms the KL divergence as a function of the privacy parameter, particularly in capturing the impact of adversarial attacks under Laplace mechanisms. Finally, we upper bound the adversary's advantage in membership inference attacks based on Chernoff DP and numerically compare its performance with existing bounds. We re-derive Chernoff DP in connection with $\varepsilon-$DP using the Radon-Nikodym derivative, and prove its relation with KL-DP. Subsequently, we present numerical evaluation results, which demonstrates that Chernoff information outperforms KL divergence as a function of the privacy parameter $\varepsilon$ and the impact of the adversary's attack in Laplace mechanisms. Lastly, we introduce a new upper bound on adversary's membership advantage in membership inference attacks using Chernoff DP and numerically compare its performance with existing alternatives based on $(\varepsilon,δ)-$DP in the literature.
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。