首页> 外文期刊>電子情報通信学会技術研究報告. 非線形問題. Nonlinear Problems >Analysis and measurement of limit cycles generated on neural networks with cyclic and longer distance connections
【24h】

Analysis and measurement of limit cycles generated on neural networks with cyclic and longer distance connections

机译:具有循环和较长距离连接的神经网络生成的极限环的分析和测量

获取原文
获取原文并翻译 | 示例
           

摘要

Analysis of the neural networks with antisymmetric and cyclic nearest connections is already reported. But the neural network with longer distance connections is not analyzed yet. In this paper, we confirm that the networks with longer distance connection generate the limit cycle similar to the nearest neighbor connections, and we study relations between limit cycle and the size of neural network, maximum distance of synapse connection. Additionally, we construct this network by means of LSI neural network chip that was designed in our laboratory and we observe the limit cycle of the LSI neural network chip.
机译:已经报道了具有反对称和循环最近连接的神经网络的分析。但是尚未分析具有更长距离连接的神经网络。在本文中,我们确认距离连接较长的网络产生的极限环与最近的邻居连接相似,并且研究了极限环与神经网络的大小,突触连接的最大距离之间的关系。此外,我们通过在实验室中设计的LSI神经网络芯片构建了该网络,并观察了LSI神经网络芯片的极限周期。

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