Nonlinear multivariate analysis based on statistical machine learning theory

2014 Fiscal Year Final Research Report

• Project/Area Number: 24700280
• Research Category: Grant-in-Aid for Young Scientists (B)
• Allocation Type: Multi-year Fund
• Research Field: Statistical science
• Research Institution: The University of Electro-Communications (2014); Osaka Prefecture University (2012-2013)
• Principal Investigator: KAWANO SHUICHI 電気通信大学, その他の研究科, 准教授 (50611448)
• Project Period (FY): 2012-04-01 – 2015-03-31
• Keywords: 半教師あり学習 / 関数データ解析 / モデル評価基準 / スパース推定 / 正則化法 / 異分布性

Outline of Final Research Achievements

We developed nonlinear statistical methods to extract useful information from high-dimensional diverse data. In particular, we proposed semi-supervised methods that can treat functional data or labeled data and unlabeled data from different sampling distributions, and developed a series of procedures for evaluating and predicting statistical models based on sparse estimation. We applied the proposed methods to datasets in the various fields of research including life science.

Free Research Field

統計科学