Parallel architecture and optimization for discrete-event simulation of spike neural networks

Spike neural networks are inspired by animal brains, and outperform traditional neural networks on complicated tasks. How- ever, spike neural networks are usually used on a large scale, and they cannot be computed on commercial, off-the-shelf com- puters. A parallel architecture is proposed and deve...

Full description

Saved in:
Bibliographic Details
Published inScience China. Technological sciences Vol. 56; no. 2; pp. 509 - 517
Main Authors Tang, YuHua, Zhang, BaiDa, Wu, JunJie, Hu, TianJiang, Zhou, Jing, Liu, FuDong
Format Journal Article
LanguageEnglish
Published Heidelberg SP Science China Press 01.02.2013
Subjects
Architecture
Computation
Computer simulation
Dynamical systems
Engineering
Neural networks
Optimization
Spikes
Tasks
仿真结果
优化
动态负载
平衡机制
并行体系结构
神经网络系统
离散事件仿真
离散事件模拟
discrete event simulation
spike neural network
intelligent parallelization framework
Online AccessGet full text

Cover

More Information
Summary:Spike neural networks are inspired by animal brains, and outperform traditional neural networks on complicated tasks. How- ever, spike neural networks are usually used on a large scale, and they cannot be computed on commercial, off-the-shelf com- puters. A parallel architecture is proposed and developed for discrete-event simulations of spike neural networks. Furthermore, mechanisms for both parallelism degree estimation and dynamic load balance are emphasized with theoretical and computa- tional analysis. Simulation results show the effectiveness of the proposed parallelized spike neural network system and its cor- responding support components.
Bibliography:spike neural network, discrete event simulation, intelligent parallelization framework
Spike neural networks are inspired by animal brains, and outperform traditional neural networks on complicated tasks. How- ever, spike neural networks are usually used on a large scale, and they cannot be computed on commercial, off-the-shelf com- puters. A parallel architecture is proposed and developed for discrete-event simulations of spike neural networks. Furthermore, mechanisms for both parallelism degree estimation and dynamic load balance are emphasized with theoretical and computa- tional analysis. Simulation results show the effectiveness of the proposed parallelized spike neural network system and its cor- responding support components.
11-5845/TH
ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 23
ISSN:1674-7321
1869-1900
DOI:10.1007/s11431-012-5084-2