Gibbs sampler optimization for analysis of a granulated medium
- Autores: Koltcov S.N.1, Nikolenko S.I.1, Koltsova E.Y.1
-
Afiliações:
- National Research University Higher School of Economics
- Edição: Volume 42, Nº 8 (2016)
- Páginas: 837-839
- Seção: Article
- URL: https://journal-vniispk.ru/1063-7850/article/view/200547
- DOI: https://doi.org/10.1134/S1063785016080241
- ID: 200547
Citar
Resumo
A new variant of the method of probability density distribution recovery for solving topical modeling problems is described. Disadvantages of the Gibbs sampling algorithm are considered, and a modified variant, called the “granulated sampling method,” is proposed. Based on the results of statistical modeling, it is shown that the proposed algorithm is characterized by higher stability as compared to other variants of Gibbs sampling.
Sobre autores
S. Koltcov
National Research University Higher School of Economics
Autor responsável pela correspondência
Email: skoltsov@hse.ru
Rússia, St. Petersburg, 190008
S. Nikolenko
National Research University Higher School of Economics
Email: skoltsov@hse.ru
Rússia, St. Petersburg, 190008
E. Koltsova
National Research University Higher School of Economics
Email: skoltsov@hse.ru
Rússia, St. Petersburg, 190008
Arquivos suplementares
