arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2601.09406cs.ITmath.IT
收藏

α-互信息的通用泄露解释

A Generalized Leakage Interpretation of Alpha-Mutual Information

Shonan Institute of Technology(湘南工科大学);Nihon University(日本大学)

Akira Kamatsuka, Takahiro Yoshida

AI总结:

本文提出了一种基于对抗性通用决策问题的框架,将α-互信息解释为对抗者风险规避程度的度量,并利用Kolmogorov-Nagumo均值和q对数来表征对抗性收益。

摘要:

本文提出了α-互信息(α-MI)在通用g-泄露框架下的统一解释。具体而言,我们基于对抗性通用决策问题扩展的定量信息流框架,提出了α-MI的新解释。该框架利用Kolmogorov-Nagumo均值和q对数来表征对抗性收益。此外,我们证明在该框架下,参数α可以被解释为对抗者风险规避程度的度量。

英文摘要:

This paper presents a unified interpretation of $α$-mutual information ($α$-MI) in terms of generalized $g$-leakage. Specifically, we present a novel interpretation of $α$-MI within an extended framework for quantitative information flow based on adversarial generalized decision problems. This framework employs the Kolmogorov-Nagumo mean and the $q$-logarithm to characterize adversarial gain. Furthermore, we demonstrate that, within this framework, the parameter $α$ can be interpreted as a measure of the adversary's risk aversion.

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。

↑