革新知能統合研究センター

speaker: Koichi Tojo

title: Description of a normalizing constant by Kirillov character formula

abstract: Parametric models are important theoretically and practically to estimate a true distribution in the field of machine learning. We can obtain a parametric model by normalizing a family of parameterized
functions on a sample space. However, in general, it is difficult to calculate the normalizing constant.
Recently, F. Barbaresco introduced a method to describe the normalizing constant of the Gibbs density on Poincaré disk by Kirillov character formula, which has been actively studied in the field of representation theory.
In this talk, we will give an overview of the method and consider its application.

time: 13:00 – 14:00 + 30 min
place: Keio Univ. Yagami-campus Bldg.14th, 6F
room: 631 A/B