摘要
利用计算神经科学原理或图论对大脑进行建模得到的大尺度大脑模型,在脑科学研究和人工智能等方面有着极大的研究意义和应用价值。合理的大尺度大脑模型将对探索和理解大脑工作的内在机制以及大脑神经系统相关疾病的成因有很大帮助,也将大大推动人工智能领域由当前的弱人工智能向强人工智能迈进。因此,大尺度大脑模型的相关研究在过去十年间受到国内外学者的广泛关注。通过查阅大量关于大尺度大脑模型的研究文献,并对其相关研究进行回顾、归纳、分析和总结,报告了大尺度大脑模型的研究现状。给出了大尺度大脑模型的明确定义,归纳总结了大尺度大脑模型的多个范畴,同时介绍了研究大尺度大脑模型所需了解的相关基础理论;归纳了大尺度大脑模型的有效构建策略,回顾了迄今为止国内外具有代表性的几个大尺度大脑模型的详细建模方法及应用;总结了大尺度大脑模拟领域目前存在的不足和遇到的困难,展望了大尺度大脑模型将来可能的发展趋势和应用方向。
The large-scale brain model which simulates brain on a basis of computational neuroscience principle or graph theory has great significance and application value in brain science research and artificial intelligence.A proper large scale brain model will contribute to exploring and understanding the underlying mechanisms of brain and the reasons of neurological diseases,and will also greatly put forward the transformation from the current weak artificial intelligence to strong one.Thus,research on large-scale brain models has received extensive attention all over the world in the past decades.By reviewing latest research literatures on large-scale brain models,this paper reports the current progress of large-scale brain models.Firstly,it gives a clear definition of large-scale brain models,outlines the categories of large-scale brain model,and introduces the relevant fundamental theories that is necessary for understanding large-scale brain models.Sec-ondly,the effective modeling strategies of large-scale brain models are summarized,and the detailed modeling methods of several representative large-scale brain models and its application are introduced.Finally,it summarizes the shortcomings and difficulties encountered in the field of large-scale brain models,and predicts the possible development and application directions of large-scale brain models.
作者
张军鹏
蒋睿
施玉杰
ZHANG Junpeng;JIANG Rui;SHI Yujie(College of Electrical Engineering,Sichuan University,Chengdu 610065,China)
出处
《计算机工程与应用》
CSCD
北大核心
2022年第14期51-62,共12页
Computer Engineering and Applications
关键词
大脑模型
大尺度模拟
人工智能
类脑计算
brain model
large-scale simulation
artificial intelligence
brain-inspired computing
