arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2403.10089cs.ITcs.CVcs.LGmath.IT
收藏

参数统计模型之间 Fisher-Rao 距离的近似与定界技术

Approximation and bounding techniques for the Fisher-Rao distances between parametric statistical models

Sony Computer Science Laboratories Inc(索尼计算机科学实验室公司)

Frank Nielsen

AI总结:

本文研究参数统计模型间 Fisher-Rao 距离的闭式计算难题,提出基于子模型、预测地线、Hessian 度量及群论不变量的近似与紧定界方法。

摘要:

统计模型中两个概率分布之间的 Fisher-Rao 距离被定义为由 Fisher 信息度量诱导的黎曼测地距离。为了以闭式形式计算 Fisher-Rao 距离,我们需要:(1)导出 Fisher-Rao 测地线的公式;(2)沿这些测地线积分 Fisher 长度元。我们考虑了若干数值稳健的 Fisher-Rao 距离近似与定界技术:首先,我们报告了基于子模型的闭式一维 Fisher-Rao 距离的通用 Fisher-Rao 距离上界。其次,我们根据 Fisher-Rao 测地线或预测地线是否能够以闭式形式获得,描述了若干通用近似方案。特别地,我们得到一种通用方法,只要能够获得 Fisher-Rao 预测地线以及紧的下界和上界,该方法就能保证近似具有任意小的加性误差。第三,我们考虑 Fisher 度量为 Hessian 度量的情形,并利用信息几何技术报告 Fisher-Rao 距离的通用紧上界。单参数和双参数统计模型总是具有 Fisher Hessian 度量,而且一般而言,可以通过一个简单检验来判断 Fisher 信息矩阵是否产生 Hessian 度量。第四,我们考虑椭圆分布族,并展示如何将上述技术应用于这些模型。我们还提出了两种新距离:一种基于充当 Fisher-Rao 测地线代理的曲线的 Fisher-Rao 长度,另一种基于 Birkhoff/Hilbert 射影锥距离。最后,我们针对统计变换模型考虑了一种基于最大不变量概念的替代群论方法,该方法为 Fisher-Rao 距离公式的结构提供了洞见,并可在应用中得到有效利用。

英文摘要:

The Fisher-Rao distance between two probability distributions of a statistical model is defined as the Riemannian geodesic distance induced by the Fisher information metric. In order to calculate the Fisher-Rao distance in closed-form, we need (1) to elicit a formula for the Fisher-Rao geodesics, and (2) to integrate the Fisher length element along those geodesics. We consider several numerically robust approximation and bounding techniques for the Fisher-Rao distances: First, we report generic upper bounds on Fisher-Rao distances based on closed-form 1D Fisher-Rao distances of submodels. Second, we describe several generic approximation schemes depending on whether the Fisher-Rao geodesics or pregeodesics are available in closed-form or not. In particular, we obtain a generic method to guarantee an arbitrarily small additive error on the approximation provided that Fisher-Rao pregeodesics and tight lower and upper bounds are available. Third, we consider the case of Fisher metrics being Hessian metrics, and report generic tight upper bounds on the Fisher-Rao distances using techniques of information geometry. Uniparametric and biparametric statistical models always have Fisher Hessian metrics, and in general a simple test allows to check whether the Fisher information matrix yields a Hessian metric or not. Fourth, we consider elliptical distribution families and show how to apply the above techniques to these models. We also propose two new distances based either on the Fisher-Rao lengths of curves serving as proxies of Fisher-Rao geodesics, or based on the Birkhoff/Hilbert projective cone distance. Last, we consider an alternative group-theoretic approach for statistical transformation models based on the notion of maximal invariant which yields insights on the structures of the Fisher-Rao distance formula which may be used fruitfully in applications.

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。

↑