Information Geometry of Reversible Markov Chains
Information Geometry of Reversible Markov Chains
复制标题
可逆马尔可夫链的信息几何
DOI:
10.1007/s41884-021-00061-7
复制
发表时间:
2021
期刊:
影响因子:
--
通讯作者:
Watanabe Shun
中科院分区:
文献类型:
--
作者:
Wolfer Geoffrey;Watanabe Shun
We analyze the information geometric structure of time reversibility for parametric families of irreducible transition kernels of Markov chains. We define and characterize reversible exponential families of Markov kernels, and show that irreducible and reversible Markov kernels form both a mixture family and, perhaps surprisingly, an exponential family in the set of all stochastic kernels. We propose a parametrization of the entire manifold of reversible kernels, and inspect reversible geodesics. We define information projections onto the reversible manifold, and derive closed-form expressions for the e-projection and m-projection, along with Pythagorean identities with respect to information divergence, leading to some new notion of reversiblization of Markov kernels. We show the family of edge measures pertaining to irreducible and reversible kernels also forms an exponential family among distributions over pairs. We further explore geometric properties of the reversible family, by comparing them with other remarkable families of stochastic matrices. Finally, we show that reversible kernels are, in a sense we define, the minimal exponential family generated by the m-family of symmetric kernels, and the smallest mixture family that comprises the e-family of memoryless kernels.
登录
查看更多内容
DOI:
--
发表时间:
--
期刊:
影响因子:
--
作者:
Adam Nielsen;Marcus Weber
通讯作者:
Marcus Weber
DOI:
--
发表时间:
--
期刊:
IEEE Transactions on Information Theory, accepted for publication
影响因子:
--
作者:
J-I. Takeuchi;A. Barron;T. Kawabata
通讯作者:
T. Kawabata
DOI:
--
发表时间:
2005
期刊:
第28回情報理論とその応用シンポジウム予稿集
影响因子:
--
作者:
Doi;T.;Fujito;T.;H.Ishihara;H.Nagaoka
通讯作者:
H.Nagaoka
DOI:
10.1109/isit.1998.708923
发表时间:
1998
期刊:
Proceedings. 1998 IEEE International Symposium on Information Theory (Cat. No.98CH36252)
影响因子:
--
作者:
Jun'ichi Takeuchi;Andrew R. Barron
通讯作者:
Andrew R. Barron
DOI:
10.1109/itw.2017.8278008
发表时间:
2017
期刊:
2017 IEEE Information Theory Workshop (ITW)
影响因子:
--
作者:
J. Takeuchi;H. Nagaoka
通讯作者:
H. Nagaoka