| Project/Area Number |
23300069
|
|---|
| Research Category |
Grant-in-Aid for Scientific Research (B)
|
| Allocation Type | Single-year Grants |
|---|
| Section | 一般 |
|---|
| Research Field |
Perception information processing/Intelligent robotics
|
|---|
| Research Institution | Tokyo Institute of Technology |
|---|
Principal Investigator |
|
|---|
| Co-Investigator(Kenkyū-buntansha) |
SUGIYAMA Masashi 東京工業大学, 大学院・情報理工学研究科, 准教授 (90334515)
TANAKA Toshihisa 東京農工大学, 大学院・工学研究院, 准教授 (70360584)
WASHIZAWA Yoshikazu 電気通信大学, 大学院・情報理工学研究科, 助教 (10419880)
|
|---|
| Project Period (FY) |
2011-04-01 – 2014-03-31
|
| Project Status |
Completed (Fiscal Year 2013)
|
| Budget Amount *help |
¥20,670,000 (Direct Cost: ¥15,900,000、Indirect Cost: ¥4,770,000)
Fiscal Year 2013: ¥4,420,000 (Direct Cost: ¥3,400,000、Indirect Cost: ¥1,020,000)
Fiscal Year 2012: ¥4,420,000 (Direct Cost: ¥3,400,000、Indirect Cost: ¥1,020,000)
Fiscal Year 2011: ¥11,830,000 (Direct Cost: ¥9,100,000、Indirect Cost: ¥2,730,000)
|
|---|
| Keywords | 時空間計量 / Mahalanobis計量 / 最小二乗確率的分類器 / 多様体上のアンサンブル学習 / マルチカーネル適応フィルタ / 制約条件付き最大事後確率識別器 / マハラノビス計量 / 拡張カーネル法 / 最小2乗確率的分類器 / 適応的カーネル主成分分析 / Fisher判別分析の修正項 / 線形計画法による最大事後確率識別 / コミッティマシン / 部分カーネル主成分追跡 / Grassmann多様体上の距離 / サポートベクタマシン |
|---|
| Research Abstract |
In order to improve the performance of signal processing and pattern recognition, it is necessary to obtain a proper metric for signals and patterns and develop processing and recognition methods based on the metric.
According to the concept, we improve the theory of Mahalanobis metric, develop a calculation method of Mahalanobis metric using kernel functions, and provide criteria and calculation methods that can obtain a equivalent result with the maximum posteriori method without estimating a posteriori probability. We also propose a method to directly estimate the L2 distance between two probability density functions and show its properties with respect to convergence and bias. We develop an algorithm to calculate the sample average on Stiefel manifold, and the adaptive kernel principle component analysis that can extract features with a small number of basis functions. The advantages of developed methods are shown by experimental results.
|
