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 2302.10070cs.ITmath.IT
收藏

Jensen-Shannon 散度幂的度量化

Metrization of powers of the Jensen-Shannon divergence

Kazuki Okamura

AI总结:

本文针对 Osán 等人关于多项分布间 Jensen-Shannon 散度分数次幂可度量化的猜想给出肯定证明,并将方法推广到 Cauchy 分布间的 $f$-散度幂。

摘要:

统计散度的度量化在理论和实践两方面都具有重要价值。获得与散度相关联的度量的一种方法是考虑其分数次幂。受这一思想启发,Osán、Bussandri 和 Lamberti(2018)研究了多项分布之间 Jensen-Shannon 散度分数次幂的度量化,并提出了一个开放问题。在这篇短札记中,我们对他们的猜想给出了肯定回答。此外,我们的方法也适用于 Cauchy 分布之间 $f$-散度的分数次幂。

英文摘要:

Metrization of statistical divergences is valuable in both theoretical and practical aspects. One approach to obtaining metrics associated with divergences is to consider their fractional powers. Motivated by this idea, Osán, Bussandri, and Lamberti (2018) studied the metrization of fractional powers of the Jensen-Shannon divergence between multinomial distributions and posed an open problem. In this short note, we provide an affirmative answer to their conjecture. Moreover, our method is also applicable to fractional powers of $f$-divergences between Cauchy distributions.

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。

↑