2010 Fiscal Year Final Research Report
Spatiotemporally organized retrieval of information inspired by the neural mechanism of mental processes
| Project/Area Number |
20500279
|
|---|
| Research Category |
Grant-in-Aid for Scientific Research (C)
|
| Allocation Type | Single-year Grants |
|---|
| Section | 一般 |
|---|
| Research Field |
Bioinformatics/Life informatics
|
|---|
| Research Institution | The Institute of Physical and Chemical Research |
|---|
Principal Investigator |
OKAMOTO Hiroshi The Institute of Physical and Chemical Research, 脳回路機能理論研究チーム, 客員研究員 (00374067)
|
|---|
| Project Period (FY) |
2008 – 2010
|
| Keywords | 脳型記憶想起 / 時間積分 / 複雑ネットワーク / 時系列想起 / ベイズ推定 |
|---|
| Research Abstract |
The purpose of this study is to develop a method of extracting temporally organized information by analogy to memory retrieval in the brain. We have demonstrated that stochastic dynamics of a grand network of recurrent neural networks, which models a minimal structure of cortical circuits, can produce transient and sequential activation of neurons. Different cue presentations represented by the initial state of grand network activation generate different temporal sequences. Statistical-mechanical exploration has revealed that retrieval of temporal sequences by this model conforms to Bayesian inference.
|
