| Project/Area Number |
16K00329
|
|---|
| Research Category |
Grant-in-Aid for Scientific Research (C)
|
| Allocation Type | Multi-year Fund |
|---|
| Section | 一般 |
|---|
| Research Field |
Soft computing
|
|---|
| Research Institution | University of Yamanashi |
|---|
Principal Investigator |
|
|---|
| Research Collaborator |
NAKANO Shunta
|
|---|
| Project Period (FY) |
2016-04-01 – 2019-03-31
|
| Project Status |
Completed (Fiscal Year 2018)
|
| Budget Amount *help |
¥3,380,000 (Direct Cost: ¥2,600,000、Indirect Cost: ¥780,000)
Fiscal Year 2018: ¥780,000 (Direct Cost: ¥600,000、Indirect Cost: ¥180,000)
Fiscal Year 2017: ¥1,170,000 (Direct Cost: ¥900,000、Indirect Cost: ¥270,000)
Fiscal Year 2016: ¥1,430,000 (Direct Cost: ¥1,100,000、Indirect Cost: ¥330,000)
|
|---|
| Keywords | 海馬 / 大脳皮質 / 記憶 / ニューラルネットワーク / 破局的忘却 / No-prop / CHL / 大脳皮質モデル / 擬似リハーサル / Hebb則 / 重みの重要度 / 擬似パターン / No-Prop / 宣言的記憶 / エピソード記憶 / 神経新生 / スパイクタイミング依存性シナプス可塑性 / ソフトコンピューティング |
|---|
| Outline of Final Research Achievements |
In this study, we were aimed at modeling the complementary learning system in the hippocampus and neocortex responsible for memories about events and facts. In particular, we focused on the neocortex, which is the locus of long-term memory. We considered what kind of mechanism would be required to prevent old memories from being destroyed by new memories, and constructed biologically plausible models. By examining their characteristics by computer simulation, we have revealed that it is possible to reduce catastrophic forgetting by combining a biologically relevant learning method that does not propagate output errors backward through the network and pseudorehearsal. We have also shown that catastrophic forgetting is further reduced by considering the importance of weights.
|
| Academic Significance and Societal Importance of the Research Achievements |
意識的に思い出すことのできる様々な出来事に関する記憶（エピソード記憶）や事実に関する記憶（意味記憶）は宣言的記憶と呼ばれ，思考や推論をといった極めて高次な情報処理で用いられている．宣言的記憶は，初めに海馬に蓄えられ，その後徐々に大脳皮質へと転写されていくと考えられているが，その仕組みは未解明である．本研究では，人間のように知的で柔軟な情報処理システムの実現に向けて，その基盤となる宣言的記憶の形成過程を工学的に模倣した．特に，人間のように，古い記憶を破壊することなく，次々と新しい情報を追加的に記憶していく仕組みについて，そのモデル化を行った．
|
