...
首页> 外文期刊>International Journal Of Modelling & Simulation >VLSI IMPLEMENTATION OF ARTIFICIAL NEURAL NETWORKS - A SURVEY
【24h】

VLSI IMPLEMENTATION OF ARTIFICIAL NEURAL NETWORKS - A SURVEY

机译:人工神经网络的VLSI实现-调查

获取原文
获取原文并翻译 | 示例
   

获取外文期刊封面封底 >>

       

摘要

Artificial neural networks (ANNs) are simplified models of human brain. These are networks of computing elements that have the ability to respond to input stimuli and generate the corresponding output. To obtain a desirable output, the network weights must be trained upon the available data many times. Hence the software realization of ANN takes many hours to learn a particular example. On the other hand, neural network (NN) in hardware can speed up the training by several orders of magnitude, due to the faster nature of the hardware. Different types of VLSI implementation of ANN are found in the literature. This paper provides a brief survey of digital and pulsed neurohardware. It highlights the important issues related and shows the possible direction of future research.
机译:人工神经网络(ANN)是人脑的简化模型。这些是具有响应输入刺激并生成相应输出的能力的计算元件网络。为了获得理想的输出,必须根据可用数据对网络权重进行多次训练。因此,ANN的软件实现需要花费大量时间来学习特定的示例。另一方面,由于硬件的更快特性,硬件中的神经网络(NN)可以将训练速度提高几个数量级。在文献中找到了不同类型的ANN的VLSI实现。本文简要概述了数字和脉冲神经硬件。它突出显示了相关的重要问题,并显示了未来研究的可能方向。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号