A comprehensive workflow for general-purpose neural modeling with highly configurable neuromorphic hardware systems

被引：61

作者：

Bruederle, Daniel ^{[1
]}

Petrovici, Mihai A. ^{[1
]}

Vogginger, Bernhard ^{[1
]}

Ehrlich, Matthias ^{[6
]}

Pfeil, Thomas ^{[1
]}

Millner, Sebastian ^{[1
]}

Gruebl, Andreas ^{[1
]}

Wendt, Karsten ^{[6
]}

Mueller, Eric ^{[1
]}

Schwartz, Marc-Olivier ^{[1
]}

de Oliveira, Dan Husmann ^{[1
]}

Jeltsch, Sebastian ^{[1
]}

Fieres, Johannes ^{[1
]}

Schilling, Moritz ^{[1
]}

Mueller, Paul ^{[1
]}

Breitwieser, Oliver ^{[1
]}

Petkov, Venelin ^{[1
]}

Muller, Lyle ^{[2
]}

Davison, Andrew P. ^{[2
]}

Krishnamurthy, Pradeep ^{[8
]}

Kremkow, Jens ^{[7
]}

Lundqvist, Mikael ^{[8
]}

Muller, Eilif ^{[9
]}

Partzsch, Johannes ^{[6
]}

Scholze, Stefan ^{[6
]}

Zuehl, Lukas ^{[6
]}

Mayr, Christian ^{[6
]}

Destexhe, Alain ^{[2
]}

Diesmann, Markus ^{[3
,4
,5
]}

Potjans, Tobias C. ^{[4
,10
]}

Lansner, Anders ^{[8
]}

Schueffny, Rene ^{[6
]}

Schemmel, Johannes ^{[1
]}

Meier, Karlheinz ^{[1
]}

机构：

[1] Heidelberg Univ, Kirchhoff Inst Phys, Heidelberg, Germany

[2] CNRS, Unite Neurosci Informat & Complexite, Gif Sur Yvette, France

[3] RIKEN, Brain Sci Inst, Wako, Saitama, Japan

[4] RIKEN, Computat Sci Res Program, Wako, Saitama, Japan

[5] Univ Freiburg, Bernstein Ctr Computat Neurosci, Freiburg, Germany

[6] Tech Univ Dresden, Inst Circuits & Syst, Dresden, Germany

[7] Univ Freiburg, Bernstein Ctr Freiburg, Freiburg, Germany

[8] KTH Stockholm, Stockholm, Sweden

[9] Ecole Polytech Fed Lausanne, Brain Mind Inst, Lausanne, Switzerland

[10] Forschungszentrum Julich, Inst Neurosci & Med INM 6, D-52425 Julich, Germany

来源：

BIOLOGICAL CYBERNETICS | 2011年 / 104卷 / 4-5期

关键词：

Neuromorphic; VLSI; Hardware; Wafer scale; Software; Modeling; Computational neuroscience; PyNN; TIMING-DEPENDENT PLASTICITY; SIGNAL PROPAGATION; SPIKING NEURONS; NETWORK MODEL; EMERGENCE; DYNAMICS; PATTERNS; ARRAY;

D O I：

10.1007/s00422-011-0435-9

中图分类号：

TP3 [计算技术、计算机技术];

学科分类号：

0812 ;

摘要：

In this article, we present a methodological framework that meets novel requirements emerging from upcoming types of accelerated and highly configurable neuromorphic hardware systems. We describe in detail a device with 45 million programmable and dynamic synapses that is currently under development, and we sketch the conceptual challenges that arise from taking this platform into operation. More specifically, we aim at the establishment of this neuromorphic system as a flexible and neuroscientifically valuable modeling tool that can be used by non-hardware experts. We consider various functional aspects to be crucial for this purpose, and we introduce a consistent workflow with detailed descriptions of all involved modules that implement the suggested steps: The integration of the hardware interface into the simulator-independent model description language PyNN; a fully automated translation between the PyNN domain and appropriate hardware configurations; an executable specification of the future neuromorphic system that can be seamlessly integrated into this biology-to-hardware mapping process as a test bench for all software layers and possible hardware design modifications; an evaluation scheme that deploys models from a dedicated benchmark library, compares the results generated by virtual or prototype hardware devices with reference software simulations and analyzes the differences. The integration of these components into one hardware-software workflow provides an ecosystem for ongoing preparative studies that support the hardware design process and represents the basis for the maturity of the model-to-hardware mapping software. The functionality and flexibility of the latter is proven with a variety of experimental results.

引用

页码：263 / 296

页数：34

共 98 条

[71] The columnar organization of the neocortex [J].

Mountcastle, VB .

BRAIN, 1997, 120 :701-722

[72] Firing patterns in the adaptive exponential integrate-and-fire model [J].

Naud, Richard ;

Marcille, Nicolas ;

Clopath, Claudia ;

Gerstner, Wulfram .

BIOLOGICAL CYBERNETICS, 2008, 99 (4-5) :335-347

[73]

Netter T, 2002, 2002 IEEE/RSJ INTERNATIONAL CONFERENCE ON INTELLIGENT ROBOTS AND SYSTEMS, VOLS 1-3, PROCEEDINGS, P129, DOI 10.1109/IRDS.2002.1041376

[74]

Norris M., 2003, GigaBit Ethernet - Techonoloy and Applications

[75] Python']Python for scientific computing [J].

Oliphant, Travis E. .

COMPUTING IN SCIENCE & ENGINEERING, 2007, 9 (03) :10-20

[76]

OSTER M, 2005, P 2005 INT C ART NEU

[77]

Pecevski D., 2009, FRONT NEUROINFORM, V3

[78] Reward-Modulated Hebbian Learning of Decision Making [J].

Pfeiffer, Michael ;

Nessler, Bernhard ;

Douglas, Rodney J. ;

Maass, Wolfgang .

NEURAL COMPUTATION, 2010, 22 (06) :1399-1444

[79]

Philipp Stefan, 2009, Proceedings 2009 International Joint Conference on Neural Networks (IJCNN 2009 - Atlanta), P2525, DOI 10.1109/IJCNN.2009.5178983

[80] Minimal Hodgkin-Huxley type models for different classes of cortical and thalamic neurons [J].

Pospischil, Martin ;

Toledo-Rodriguez, Maria ;

Monier, Cyril ;

Piwkowska, Zuzanna ;

Bal, Thierry ;

Fregnac, Yves ;

Markram, Henry ;

Destexhe, Alain .

BIOLOGICAL CYBERNETICS, 2008, 99 (4-5) :427-441

← 1 2 3 4 5 6 7 8 9 10 →