| Project/Area Number |
05680301
|
|---|
| Research Category |
Grant-in-Aid for General Scientific Research (C)
|
| Allocation Type | Single-year Grants |
|---|
| Research Field |
Intelligent informatics
|
|---|
| Research Institution | Kyushu Institute of Technogy |
|---|
Principal Investigator |
MATSUOKA Kiyotoshi Department of Control Engineering Kyushu Institute of Technology, Professor, 工学部, 教授 (90110840)
|
|---|
| Co-Investigator(Kenkyū-buntansha) |
TOKUNARI Tsuyoshi Department of Control Engineering, Kyushu Institute of Technology, Assistant Pro, 工学部, 助手 (00237075)
|
|---|
| Project Period (FY) |
1993 – 1994
|
| Project Status |
Completed (Fiscal Year 1994)
|
| Budget Amount *help |
¥1,100,000 (Direct Cost: ¥1,100,000)
Fiscal Year 1994: ¥400,000 (Direct Cost: ¥400,000)
Fiscal Year 1993: ¥700,000 (Direct Cost: ¥700,000)
|
|---|
| Keywords | Neural net / Hebbian rule / Anti-Hebbian rule / Principal component / Blind signal separation / 相関 / Hebb学習 / 反Hebb学習 / ニュートラルネット / 学習理論 / blind separation |
|---|
| Research Abstract |
This project was to establish a theory of anti-Hebbian learning. First, A neural network that self-organizes based on Hebbian and anti-Hebbian rules was investigated mathematically. As a result it was proved that three functions related to principal component analysis can be acquired by the proposed model with the different set of parameters. Next it was clarified that the model can also be applied to the so-called blind signal separation. Some computer simulations were also performed to make sure the mathematical result.
|
