106 0

Graph Neural Networks for Connectivity Inference in Spatially Patterned Neural Responses

Title
Graph Neural Networks for Connectivity Inference in Spatially Patterned Neural Responses
Author
박태훈
Advisor(s)
윤기중
Issue Date
2023. 2
Publisher
한양대학교
Degree
Master
Abstract
A continuous attractor network is one of the most common theoretical framework for studying a wide range of neural computations in the brain. Many previous approaches have attempted to identify continuous attractor systems by investigating the state-space structure of population neural activity. However, establishing the patterns of connectivity for relating the structure of attractor networks to their function is still an open problem. In this work, we propose the use of graph neural networks combined with the structure learning for inferring the recurrent connectivity of a ring attractor network and demonstrate that the developed model greatly improves the quality of circuit inference as well as the prediction of neural responses compared to baseline inference algorithms. |연속 어트랙터 네트워크는 (Continuous Attractor Network) 뇌의 광범위한 신경 연산을 연구할 때 가장 많이 사용되는 이론적 구조 중 하나이다. 이전에는 신경 활동 빈도의 상태-공간 (state-space) 구조를 조사하여 어트랙터 네트워크를 연구해왔다. 그러나 해당 네트워크의 구조와 기능에 대한 패턴을 정립하는 것은 여전한 문제로 남아있다. 본 연구에서, 우리는 그래프 뉴럴 네트워크를 (Graph Neural Network) 사용하여 어트랙터 네트워크의 구조와 반복적인 연결성을 추론하는 모델을 제안한다. 해당 모델을 통한 신경 반응의 예측과 뇌의 신경회로 추론의 품질이 크게 향상된 것을 확인했다.
URI
http://hanyang.dcollection.net/common/orgView/200000652878https://repository.hanyang.ac.kr/handle/20.500.11754/179692
Appears in Collections:
GRADUATE SCHOOL[S](대학원) > DEPARTMENT OF ELECTRONIC ENGINEERING(융합전자공학과) > Theses (Master)
Files in This Item:
There are no files associated with this item.
Export
RIS (EndNote)
XLS (Excel)
XML


qrcode

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.

BROWSE