arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1608.03979math.PRcs.ITmath.DGmath.FAmath.ITmath.STstat.TH
收藏

可微密度流形

Manifolds of Differentiable Densities

Nigel J. Newton

摘要:

我们开发了一个家庭的无限维(非参数)流形的概率测度。后者定义在基本的Banach空间,并有密度类$C_b^k$关于适当的参考措施。的情况下,$k=\infty$,其中的流形上的Fréchet空间建模,包括在内。流形承认费舍尔-拉奥度量,并不寻常的非参数设置,阿马里的$α$ -共变导数的所有$α\in R$。通过构造,它们是有限测度特殊流形的$C^\infty$ -嵌入子流形.统计流形是对偶($α=\pm 1$)平坦的,并承认混合和指数表示为图表。导出了它们关于$α$ -共变导数的曲率。与有限样本相关联的似然函数是每个流形上的连续函数,并且$α$发散属于$C^\infty$类。

英文摘要:

We develop a family of infinite-dimensional (non-parametric) manifolds of probability measures. The latter are defined on underlying Banach spaces, and have densities of class $C_b^k$ with respect to appropriate reference measures. The case $k=\infty$, in which the manifolds are modelled on Fréchet spaces, is included. The manifolds admit the Fisher-Rao metric and, unusually for the non-parametric setting, Amari's $α$-covariant derivatives for all $α\in R$. By construction, they are $C^\infty$-embedded submanifolds of particular manifolds of finite measures. The statistical manifolds are dually ($α=\pm 1$) flat, and admit mixture and exponential representations as charts. Their curvatures with respect to the $α$-covariant derivatives are derived. The likelihood function associated with a finite sample is a continuous function on each of the manifolds, and the $α$-divergences are of class $C^\infty$.

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。

↑