| 小西 卓哉 | 1261005: D, 中間発表 | 池田 和司, 松本 裕治, 渡辺 一帆, 久保 孝富 |
|
title: A study of variational inference algorithm for Bayesian relational model
abstract: To analyze relational data which appear in many fields of natural science, statistical models have been widely used. We focus on the Infinite relational model (IRM) which is a Bayesian relational model and provides a latent cluster structure between objects. In previous studies, IRM applied relatively small datasets. To adapt large-scale datasets and practical situations, we consider variational inference algorithms which are deterministic and possible to assess convergence. We derived some variational inference algorithms for IRM, and validate these algorithms. We further show the future works under consideration. language of the presentation: Japanese | ||
| 椿 真史 | 1251067: M, 2回目発表 | 松本 裕治, 池田 和司, 新保 仁, Kevin Duh |
| 和良品 友大 | 1251121: M, 2回目発表 | 松本 裕治, 池田 和司, 山田 武士, 新保 仁 |
| 田村 潤 | 1251062: M, 2回目発表 | 池田 和司, 松本 裕治, 久保 孝富 |