2.793

2018影响因子

(CJCR)

  • 中文核心
  • EI
  • 中国科技核心
  • Scopus
  • CSCD
  • 英国科学文摘

留言板

尊敬的读者、作者、审稿人, 关于本刊的投稿、审稿、编辑和出版的任何问题, 您可以本页添加留言。我们将尽快给您答复。谢谢您的支持!

姓名
邮箱
手机号码
标题
留言内容
验证码

大规模类脑计算系统BiCoSS: 架构、实现及应用

杨双鸣 郝新宇 王江 李会艳 魏熙乐 于海涛 邓斌

杨双鸣, 郝新宇, 王江, 李会艳, 魏熙乐, 于海涛, 邓斌. 大规模类脑计算系统BiCoSS: 架构、实现及应用. 自动化学报, 2021, 47(9): 1−16 doi: 10.16383/j.aas.c190035
引用本文: 杨双鸣, 郝新宇, 王江, 李会艳, 魏熙乐, 于海涛, 邓斌. 大规模类脑计算系统BiCoSS: 架构、实现及应用. 自动化学报, 2021, 47(9): 1−16 doi: 10.16383/j.aas.c190035
Yang Shuang-Ming, Hao Xin-Yu, Wang Jiang, Li Hui Yan, Wei Xi-Le, Yu Hai-Tao, Deng Bin. Large-scale brain-inspired computing system biCoSS: its architecture, implementation and application. Acta Automatica Sinica, 2021, 47(9): 1−16 doi: 10.16383/j.aas.c190035
Citation: Yang Shuang-Ming, Hao Xin-Yu, Wang Jiang, Li Hui Yan, Wei Xi-Le, Yu Hai-Tao, Deng Bin. Large-scale brain-inspired computing system biCoSS: its architecture, implementation and application. Acta Automatica Sinica, 2021, 47(9): 1−16 doi: 10.16383/j.aas.c190035

大规模类脑计算系统BiCoSS: 架构、实现及应用

doi: 10.16383/j.aas.c190035
基金项目: 国家自然科学基金(61871287, 61671320, 61601320, 61771330)62071324, 62006170, 天津市自然科学基金(18JCZDJC32000)资助,中国博士后科学基金(2020M680885),
详细信息
    作者简介:

    杨双鸣:天津大学电气自动化与信息工程学院博士研究生. 主要研究方向为类脑智能与神经计算. E-mail: yangshuangming@tju.edu.cn

    郝新宇:天津大学电气自动化与信息工程学院博士研究生. 主要研究方向为神经计算及FPGA实现. E-mail: haoxy@tju.edu.cn

    王江:天津大学电气自动化与信息工程学院教授. 主要研究方向为神经控制工程与神经科学. E-mail: jiangwang@tju.edu.cn

    李会艳:天津职业技术师范大学自动化与电气工程学院教授. 主要研究方向为非线性系统与神经网络. E-mail: lhy2740@126.com

    魏熙乐:天津大学电气自动化与信息工程学院教授. 主要研究方向为神经控制工程与无创式脑调制技术. E-mail: xilewei@tju.edu.cn

    于海涛:天津大学电气自动化与信息工程学院副教授. 主要研究方向为神经系统建模与动力学分析. E-mail: htyu@tju.edu.cn

    邓斌:天津大学电气自动化与信息工程学院教授. 主要研究方向为神经计算及其非线性动力学分析. E-mail: dengbin@tju.edu.cn

  • 收稿日期 2019-01-14 录用日期 2019-06-06 Manuscript received January 14, 2019; accepted June 6, 2019 国家自然科学基金(61871287, 61671320, 61601320, 61771330), 62071324, 62006170 天津市自然科学基金 (18JCZDJC32000) 资助中国博士后科学基金(2020M680885) Supported by National Natural Science Foundation of China (61871287, 61671320, 61601320, 61771330) 62071324, 62006170, China Postdoctoral Science Foundation (Grant No. 2020M680885), Natural Science Foundation of Tianjin, China (18JCZDJC32000) 本文责任编委 曾志刚 Recommended by Associate Editor ZENG Zhi-Gang 1. 天津大学电气自动化与信息工程学院 天津 300072 2. 天津职业技术师范大学自动化与电气工程学院 天津 300222 1. School of Electrical and Information Engineering, TianjinUniversity, Tianjin 300072 2. School of Automation and Elec-
  • trical Engineering, Tianjin University of Technology and Educations, Tianjin 300222

Large-scale Brain-inspired Computing System BiCoSS: Its Architecture, Implementation and Application

Funds: Supported by National Natural Science Foundation of China (61871287, 61671320, 61601320, 61771330,62071324,62006170), Cina Postdoctoral Science Foundation (Grant No. 2020M680885), Natural Science Foundation of Tianjin, China (18JCZDJC32000)
More Information
    Author Bio:

    YANG Shuang-Ming Ph.D. candidate at the School of Electrical and Information Engineering, Tianjin University. His research interest covers brain-inspired intelligence and neural computing

    HAO Xin-Yu Ph.D. candidate at the School of Electrical and Information Engineering, Tianjin University. His research interest covers neural computing and its FPGA implementation

    WANG Jiang Professor at the School of Electrical and Information Engineering, Tianjin University. His research interest covers neural control engineering and neuroscience

    LI Hui-Yan Professor at the School of Automation and Electrical Engineering, Tianjin University of Technology and Education. Her research interest covers nonlinear systems and neural networks

    WEI Xi-Le Professor at the School of Electrical and Information Engineering, Tianjin University. His research interest covers neural control engineering and noninvasive brain modulation technology

    YU Hai-Tao Associate professor at the School of Electrical and Information Engineering, Tianjin University. His research interest covers neural system modeling and dynamics analysis

    DENG Bin Professor at the School of Electrical and Information Engineering, Tianjin University. His research interest covers neural computing and nonlinear dynamics analysis

  • 摘要: 人脑具有协同多种认知功能的能力与极强的自主学习能力, 随着脑与神经科学的快速发展, 亟需计算结构模拟人脑的、性能更强大的计算平台进行人脑智能与认知行为机制的进一步探索. 受人脑神经机制的启发, 本文提出了基于神经认知计算架构的众核类脑计算系统BiCoSS, 该系统以并行计算的现场可编程门阵列(Field-programmable gate array, FPGA)为核心处理器, 以地址事件表达的神经放电作为信息传递载体, 以具有认知计算功能的神经元作为信息处理单元, 实现了四百万神经元数量级大规模神经元网络认知行为的实时计算, 填补了从细胞动力学层面理解人脑认知功能的鸿沟. 实验结果从计算能力、计算效率、功耗、通信效率、可扩展性等方面显示了BiCoSS系统的优越性能. BiCoSS通过人脑信息处理的计算架构以更贴近神经科学本质的模式实现了类脑智能; 同时, BiCoSS为神经认知和类脑计算的研究和应用提供了新的有效手段.
    1)  收稿日期 2019-01-14 录用日期 2019-06-06 Manuscript received January 14, 2019; accepted June 6, 2019 国家自然科学基金(61871287, 61671320, 61601320, 61771330), 62071324, 62006170 天津市自然科学基金 (18JCZDJC32000) 资助中国博士后科学基金(2020M680885) Supported by National Natural Science Foundation of China (61871287, 61671320, 61601320, 61771330) 62071324, 62006170, China Postdoctoral Science Foundation (Grant No. 2020M680885), Natural Science Foundation of Tianjin, China (18JCZDJC32000) 本文责任编委 曾志刚 Recommended by Associate Editor ZENG Zhi-Gang 1. 天津大学电气自动化与信息工程学院 天津 300072 2. 天津职业技术师范大学自动化与电气工程学院 天津 300222 1. School of Electrical and Information Engineering, TianjinUniversity, Tianjin 300072 2. School of Automation and Elec-
    2)  trical Engineering, Tianjin University of Technology and Educations, Tianjin 300222
  • 图  1  BiCoSS系统架构

    Fig.  1  System architecture of BiCoSS

    图  2  BiCoSS系统实物图

    Fig.  2  Physical map of BiCoSS

    图  3  BiCoSS系统神经元网络计算架构

    Fig.  3  Neural network computing architecture of BiCoSS system

    图  4  BiCoSS系统神经元与突触计算模块架构

    Fig.  4  Neuron and synapse computing architecture of BiCoSS system

    图  5  生物启发的神经元网络放电行为与突触可塑性

    Fig.  5  Spiking activities of biologically inspired neuron model and STDP characteristics

    图  6  BiCoSS系统的神经放电信息路由

    Fig.  6  Spike information routing of BiCoSS system

    图  7  模型实现的性能分析

    Fig.  7  Analysis performance of model implementation

    图  8  BiCoSS系统性能分析

    Fig.  8  Performance analysis of BiCoSS system

    图  9  BiCoSS系统神经元网络单元平均延迟

    Fig.  9  Average latency of neural network unit on BiCoSS

    图  10  实验系统实物图与计算结果

    Fig.  10  Experimental setup of BiCoSS system and computing results

    图  11  基于BiCoSS系统的认知计算

    Fig.  11  Cognition computing based on BiCoSS system

    表  1  当前路由器相关地址编码

    Table  1  The address coding of the current router

    当前节点子节点邻居节点父节点负责神经计算单元
    001, 0000001, 0100010, 00000000; 0001, 0010; 0011
    001, 0100001, 0100010, 00000000; 0001, 0010; 0011
    001, 1000001, 0100010, 00000000; 0001, 0010; 0011
    001, 1100001, 0100010, 00000000; 0001, 0010; 0011
    010, 0000001, 0100010, 00000000; 0001, 0010; 0011
    010, 1000001, 0100010, 00000000; 0001, 0010; 0011
    下载: 导出CSV

    表  2  与当前代表性大规模类脑计算系统比较

    Table  2  The comparison with the state-of-the art large-scale brain-inspired computing systems

    类脑计算系统实现模型学习规模扩展性
    BrainScaleS[24]模拟AIFSTDP4 MN2
    Truenorth[5]数字LIF1 MN2
    Neurogrid[22]混合QIF1 M2$^{N}$
    SpiNNaker[21]数字任意STDP1 BN2
    LaCSNN[28]数字任意STDP1 MN2
    BlueHive[31]数字任意64 kN2
    IFAT[32]模拟LIF65 k2$^{N}$
    HiAER[29]模拟LIF1 M2$^{N}$
    BiCoSS 数字任意STDP4 M4$\cdot $2$^{N}$
    下载: 导出CSV

    表  3  基底核模型中不同细胞的参数值

    Table  3  Parameter values of different cells in the basal ganglia model

    参数GPeGPiSTN
    $a$0.10.10.005
    $b$0.20.20.265
    $c$−65−65−65
    $d$221.5
    $I^{x}$(nA)101030
    $E_{ {\rm{AMPA} } }$(mV)000
    $E_{ {\rm{NMDA} } }$(mV)000
    $E_{ {\rm{GABA} } }$(mV)−60−60−60
    ${\tau }_{ {\rm{AMPA} } }$(ms)666
    ${\tau }_{{\rm{NMDA}}}$(ms)160160160
    ${\tau }_{{\rm{GABA}}}$(ms)444
    $W_{{\rm{Str}}D2\to GPe}$0.8
    $W_{{\rm{Str}}D1\to GPi}$1
    $W_{{\rm{STN}}\to GPi}$1.15
    下载: 导出CSV

    表  4  小脑模型中不同细胞的参数值

    Table  4  Parameter values of different cerebellar cells

    GRGOPCBSVNIO
    $\theta $(mV)−35−52−55−55−38−50
    $C$(pF)3.128106107122.310
    $G_{{\rm{leak}}}$(nS)0.432.32.322.321.630.67
    $E_{{\rm{leak}}}$(mV)−58−55−68−68−56−60
    $G_{{\rm{exc1}}}$(nS)0.158436.411331
    $G_{{\rm{exc2}}}$(nS)0.02163.00317
    $G_{{\rm{exc3}}}$(nS)6.097
    ${\tau }_{{\rm{exc1}}}$(ms)1.21.58.38.39.910
    ${\tau }_{{\rm{exc2}}}$(ms)523130.6
    ${\tau }_{{\rm{exc3}}}$(ms)170
    $E_{{\rm{exc}}}$(mV)000000
    $G_{{\rm{inh1}}}$(nS)0.0121300.18
    $G_{{\rm{inh2}}}$(nS)0.016
    ${\tau }_{{\rm{inh1}}}$(ms)71042.310
    ${\tau }_{{\rm{inh2}}}$(ms)59
    $E_{{\rm{inh}}}$(mV)−82−75−88−75
    $G_{{\rm{ahp}}}$(nS)1201001501
    $E_{{\rm{ahp}}}$(mV)−82−72.7−70−70−70−75
    ${\tau }_{{\rm{ahp}}}$(ms)5552.52.510
    $I$(nA)250700
    下载: 导出CSV

    表  5  皮层−基底核−丘脑皮层模型中不同神经元的参数值

    Table  5  Parameter values of different cells in the cortico-basal ganglia-thalamocortical model

    $a$$b$$c$$d$$I_{{\rm{app}}}$(pA)
    GPe0.0050.585−65410
    GPi0.0050.585−65410
    STN0.0060.262−6525
    TC0.0020.2−6520
    下载: 导出CSV

    表  6  皮层−基底核−丘脑皮层模型网络连接权重

    Table  6  Parameter values of synaptic coupling weight in the cortico-basal ganglia-thalamocortical model

    源节点$\to $目的节点突触连接权重$g_{ij}$
    GPe$\to $GPe0.075 + $g_{{\rm{inc}}}$
    GPe$\to $STN0.025 + $g_{{\rm{inc}}}$
    GPe$\to $GPi0.015 + $g_{{\rm{inc}}}$
    STN$\to $GPe0.075 + $g_{{\rm{inc}}}$
    STN$\to $GPi0.01 + 5$g_{ {\rm{inc} } }$
    GPi$\to $TC0.01 + 5$g_{ {\rm{inc} } }$
    下载: 导出CSV
  • [1] 蒲慕明, 徐波, 谭铁牛. 脑科学与类脑研究概述. 中国科学院院刊, 2016, 31(7): 725−736

    Poo Mu-Ming, Xu Bo, Tan Tie-Niu. Brain science and brainInspired intelligence technology-an overview. Bulletin of Chinese Academy of Sciences, 2016, 31(7): 725−736
    [2] 徐波, 刘成林, 曾毅. 类脑智能研究现状与发展思考. 中国科学院院刊, 2016, 31(7): 793−802

    Xu Bo, Liu Cheng-Lin, Zeng Yi. Research status and developments of brain-inspired intelligence. Bulletin of Chinese Academy of Sciences, 2016, 31(7): 793−802
    [3] 王力为, 许丽, 徐萍, 于汉超, 孔明辉, 沈毅, 张永清. 面向未来的中国科学院脑科学与类脑智能研究—强化基础研究, 推进深度融合. 中国科学院院刊, 2016, 31(7): 747−754

    Wang Li-Wei, Xu Li, Xu Ping, Yu Han-Chao, Kong Ming-Hui, Shen Yi, Zhang Yong-Qing. Brain science and brain-like intelligence research in Chinese academy of sciences. Bulletin of Chinese Academy of Sciences, 2016, 31(7): 747−754
    [4] Park H J, Friston K. Structural and functional brain networks: from connections to cognition. Science, 2013, 342(6158): 1238411 doi: 10.1126/science.1238411
    [5] Merolla P A, Arthur J V, Alvarez-Icaza R, Cassidy A S, Sawada J, Akopyan F, et al. A million spiking-neuron integrated circuit with a scalable communication network and interface. Science, 2014, 345(6197): 668−673 doi: 10.1126/science.1254642
    [6] Izhikevich E M, Edelman G M. Large-scale model of mammalian thalamocortical systems. Proceedings of the National Academy of Sciences, 2008, 105(9): 3593−3598 doi: 10.1073/pnas.0712231105
    [7] Yang S M, Wei X L, Wang J, Deng B, Liu C, Yu H T, Li H Y. Efficient hardware implementation of the subthalamic nucleus-external globus pallidus oscillation system and its dynamics investigation. Neural Networks, 2018, 94: 220−238
    [8] Yang S M, Deng B, Li H Y, Liu C, Wang J, Yu H T, Qin Y M. FPGA implementation of hippocampal spiking network and its real-time simulation on dynamical neuromodulation of oscillations. Neurocomputing, 2018, 282: 262−276 doi: 10.1016/j.neucom.2017.12.031
    [9] 王晓峰, 杨亚东. 基于生态演化的通用智能系统结构模型研究. 自动化学报, 2020, 46(5): 1017−1030 doi: 10.3969/j.issn.1003-8930.2019.01.001

    Wang Xiao-Feng, Yang Ya-Dong. Research on structure model of general intelligent system based on ecological evolution. Acta Automatica Sinica, 2020, 46(5): 1017−1030 doi: 10.3969/j.issn.1003-8930.2019.01.001
    [10] Neckar A, Fok S, Benjamin B V, Stewart T C, Oza N N, Voelker A R, et al. Braindrop: A mixedsignal neuromorphic architecture with a dynamical systemsbased programming model. Proceedings of the IEEE, 2019, 107(1): 144−164 doi: 10.1109/JPROC.2018.2881432
    [11] 张慧, 王坤峰, 王飞跃. 深度学习在目标视觉检测中的应用进展与展望. 自动化学报, 2017, 43(8): 1289−1305

    Zhang Hui, Wang Kun-Feng, Wang Fei-Yue. Advances and perspectives on applications of deep learning in visual object detection. Acta Automatica Sinica, 2017, 43(8): 1289−1305
    [12] 陈伟宏, 安吉尧, 李仁发, 李万里. 深度学习认知计算综述. 自动化学报, 2017, 43(11): 1886−1997

    Chen Wei-Hong, An Ji-Yao, Li Ren-Fa, Li Wan-Li. Review on deep-learning-based cognitive computing. Acta Automatica Sinica, 2017, 43(11): 1886−1997
    [13] Hassabis D, Kumaran D, Summerfield C, Botvinick M. Neuroscience-inspired artificial intelligence. Neuron, 2017, 95(2): 245−258 doi: 10.1016/j.neuron.2017.06.011
    [14] Thakur C S, Molin J L, Cauwenberghs G, Indiveri G, Kumar K, Qian N, et al. Large-scale neuromorphic spiking array processors: A quest to mimic the brain. Frontiers in Neuroscience, 2018, 12: 891 doi: 10.3389/fnins.2018.00891
    [15] Eliasmith C, Trujillo O. The use and abuse of large-scale brain models. Current Opinion in Neurobiology, 2014, 25: 1−6 doi: 10.1016/j.conb.2013.09.009
    [16] Izhikevich E M, Edelman G M. Large-scale model of mammalian thalamocortical systems. Proceedings of the National Academy of Sciences, 2008, 105(9): 3593−3598 doi: 10.1073/pnas.0712231105
    [17] Yang S M, Deng B, Wang J, Li H Y, Lu M L, Che Y Q, et al. Scalable digital neuromorphic architecture for large-scale biophysically meaningful neural network with multi-compartment neurons. IEEE Transactions on Neural Networks and Learning Systems, 2020, 31(1): 148−162
    [18] Markram H. The blue brain project. Nature Reviews Neuroscience, 2006, 7(2): 153 doi: 10.1038/nrn1848
    [19] Markram H. The human brain project. Scientific American, 2012, 306(6): 50−55 doi: 10.1038/scientificamerican0612-50
    [20] Insel T R, Landis S C, Collins F S. The NIH brain initiative. Science, 2013, 340(6133): 687−688 doi: 10.1126/science.1239276
    [21] Furber S B, Galluppi F, Temple S, Plana L A. The SpiNNaker Project. Proceedings of the IEEE, 2014, 102(5): 652−665 doi: 10.1109/JPROC.2014.2304638
    [22] Benjamin B V, Gao P, McQuinn E, Choudhary S, Chandrasekaram A R, Bussat J M. Neurogrid: A mixed-analog-digital multichip system for large-scale neural simulations. Proceedings of the IEEE, 2014, 102(5): 699−716 doi: 10.1109/JPROC.2014.2313565
    [23] Kumar S. Introducing qualcomm zeroth processors: Brain-inspired computing. Qualcomm OnQ Blog, 2013: 1−11
    [24] Petrovici M A, Vogginger B, Müller P,Breitwieser O, Lundqvist M, Muller L, et al. Characterization and compensation of network-level anomalies in mixed-signal neuromorphic modeling platforms. PloS one, 2014, 9(10): e108590 doi: 10.1371/journal.pone.0108590
    [25] Chen T S, Du Z D, Sun N H, Wang J, Wu C Y, Chen Y J, Temam O. Diannao: A small-footprint high-throughput accelerator for ubiquitous machinelearning. ACM Sigplan Notices, 2014, 49(4): 269−284
    [26] Chen Y J, Luo T, Liu S L, Zhang S J, He L Q, Wang J, et al. Dadiannao: A machinelearning supercomputer. In: Proceedings of the 47th Annual IEEE/ACM International Symposium on Microarchitecture, Cambridge, UK: IEEE, 2014. 609−622
    [27] Song S, Miller K D, Abbott L F. Competitive Hebbian learning through spike-timing-dependent synaptic plasticity. Nature Neuroscience, 2000, 3(9): 919−926 doi: 10.1038/78829
    [28] Yang S M, Wang J, Deng B, Liu C, Li H Y, Fietkiewicz C, et al. Real-time neuromorphic system for large-scale conductance-based spiking neural networks. IEEE Transactions on Cybernetics, 2019, 49(7): 2490−2503
    [29] Park J, Yu T, Joshi S, Maier C, Cauwenberghs G. Hierarchical address event routing for reconfigurable large-scale neuromorphic systems. IEEE Transactions on Neural Networks and Learning Systems, 2017, 28(10): 2408−2422 doi: 10.1109/TNNLS.2016.2572164
    [30] Pinsky P F, Rinzel J. Synchrony measures for biological neural networks. Biological Cybernetics, 1995, 73(2): 129−137 doi: 10.1007/BF00204051
    [31] Moore S W, Fox P J, Marsh S J T, Markettos A T, Mujumdar A. Bluehive—A field-programmable custom computing machine for extremescale real-time neural network simulation. In: Proceedings of the 20th IEEE International Symposium on Field-Programmable Custom Computing Machines. Toronto, ON, Canada: IEEE, 2012. 133−140
    [32] Yu T, Park J, Joshi S, Maier C, Cauwenberghs G. 65k-neuron integrate-and-fire array transceiver with address-event reconfigurable synaptic routing. In: Proceedings of the 2012 IEEE Biomedical Circuits and Systems Conference (BioCAS), Hsinchu, China, 2012. 21−24
    [33] Yamazaki T, Tanaka S. A spiking network model for passage-of-time representation in the cerebellum. European Journal of Neuroscience, 2007, 26(8): 2279−2292 doi: 10.1111/j.1460-9568.2007.05837.x
  • 加载中
图(11) / 表(6)
计量
  • 文章访问数:  158
  • HTML全文浏览量:  32
  • PDF下载量:  31
  • 被引次数: 0
出版历程
  • 收稿日期:  2019-01-14
  • 录用日期:  2019-06-06
  • 网络出版日期:  2021-07-26

目录

    /

    返回文章
    返回