Presentation 2014-07-18
Geometry of Markov Chains, Finite State Machines, and Tree Models
Jun'ichi TAKEUCHI,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) We discuss information geometrical aspects of models of Markov chains. It is known that the models of Markov chains defined by a strongly connected graph are exponential families in asymptotic sense. In this manuscript, we try to understand this fact in the view point of the embedding exponential curvature (e-curvature for short). In particular, we show that the e-curvature of the models of Markov chains defined by a strongly connected graph is of order O(1/n), where n is the data size.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) exponential family / Markov model / exponential connection
Paper # IT2014-40
Date of Issue

Conference Information
Committee IT
Conference Date 2014/7/10(1days)
Place (in Japanese) (See Japanese page)
Place (in English)
Topics (in Japanese) (See Japanese page)
Topics (in English)
Chair
Vice Chair
Secretary
Assistant

Paper Information
Registration To Information Theory (IT)
Language ENG
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Geometry of Markov Chains, Finite State Machines, and Tree Models
Sub Title (in English)
Keyword(1) exponential family
Keyword(2) Markov model
Keyword(3) exponential connection
1st Author's Name Jun'ichi TAKEUCHI
1st Author's Affiliation Faculty of Information Science and Electrical Engineering, Kyushu University()
Date 2014-07-18
Paper # IT2014-40
Volume (vol) vol.114
Number (no) 138
Page pp.pp.-
#Pages 6
Date of Issue