2018 Fiscal Year Final Research Report
computational role of neural synchronization
| Project/Area Number |
16K00409
|
|---|
| Research Category |
Grant-in-Aid for Scientific Research (C)
|
| Allocation Type | Multi-year Fund |
|---|
| Section | 一般 |
|---|
| Research Field |
Life / Health / Medical informatics
|
|---|
| Research Institution | Fukuoka Institute of Technology |
|---|
Principal Investigator |
|
|---|
| Co-Investigator(Kenkyū-buntansha) |
奈良 重俊 岡山大学, 自然科学研究科, 特命教授 (60231495)
|
|---|
| Research Collaborator |
Tsuda Ichiro
|
|---|
| Project Period (FY) |
2016-04-01 – 2019-03-31
|
| Keywords | 分節化 / 同期振動 / カオス / レザバー計算 |
|---|
| Outline of Final Research Achievements |
Using reservoir computing approach, we successfully developed a clustering method of nonstationary time series using recurrent neural networks.
To investigate information transmission using oscillatory dynamics, we developed two neural network models described below. 1) We developed a spiking neural network model of transmitting and decoding multiplexed signals using chaotic spike trains. By introducing background oscillation and phase-shift of individual chaotic neurons, we show that transmitting and decoding of multiple signals through one channel is possible. 2) We study multiple information through chaotic neural networks. We show that chaotic dynamics can contribute to the functional robustness of the networks.
|
| Free Research Field |
計算論的神経科学
|
| Academic Significance and Societal Importance of the Research Achievements |
本研究において離点間同期通信，情報の多重伝搬，同時デコーディングという脳的機能の特徴の一部を実現し，情報処理機構解明研究への寄与とした．このメカニズムには同期振動やカオス，位相差，カオス的遍歴，等の非線形ダイナミクスが重要な役割を果たしていることがわかった．この成果により非線形振動現象を通じた脳機能理解や人工知能分野へにおいて役割を果たしたと考えられる．
|
