1. [Home]
  2. [Research achievement]
  3. [Research achievement detail]

Research achievement detail

Title A Note on the Asymptotics of Bayes Prediction for Hierarchical Models (in Japanese)
Authors Takeo Takumi 、Tota Suko 、Toshiyasu Matsushima
Released Year 2004
Format Conference
Category Source coding
Jounal Name Proceedings of the 27th Symposium on Information Theory and Its Applications
Jounal Page vol.2, pp.639-642
Published Year 2004
Published Month 12
Abstract
(English)
In this paper, we consider bayes prediction for hierarchical models including gaussian mixture model, hidden Markov model, and so on.We derive the bayes predictive distribution weighted by posterior probability of models and evaluate the asymptotics of its cumulative loss and risk.
Note
(English)
1
Manuscript
Presentation