Title :
Information geometry of statistical inference - an overview
Author :
Amari, Shun-Ichi
Author_Institution :
Brain Sci. Inst., RIKEN, Wako, Japan
Abstract :
The present paper gives a short introduction to information geometry, by using a simple model of an exponential family which is a dually flat Riemannian space. The paper then overviews some of the applications of information geometry: 1) the higher-order asymptotic theory of estimation; 2) semiparametric estimation of the parameter of interest; 3) learning neural networks under the Riemannian structure; and 4) analysis of turbo codes, low density parity check codes and belief propagation algorithm.
Keywords :
belief maintenance; inference mechanisms; information theory; learning (artificial intelligence); neural nets; parameter estimation; parity check codes; statistical analysis; turbo codes; LDPC codes; Riemannian structure; belief propagation algorithm; dually flat Riemannian space; higher-order asymptotic theory; information geometry; learning neural networks; low density parity check codes; semiparametric estimation; statistical inference; turbo codes; Algorithm design and analysis; Belief propagation; Estimation theory; Information analysis; Information geometry; Neural networks; Parameter estimation; Parity check codes; Solid modeling; Turbo codes;
Conference_Titel :
Information Theory Workshop, 2002. Proceedings of the 2002 IEEE
Print_ISBN :
0-7803-7629-3
DOI :
10.1109/ITW.2002.1115423
