| 联想记忆系统学习算法的改进 |
| |
| 引用本文： | 邢春峰,王信峰.联想记忆系统学习算法的改进[J].数学的实践与认识,2004,34(9):92-96. |
| |
| 作者姓名： | 邢春峰 王信峰 |
| |
| 作者单位： | 北京联合大学基础部,北京,100101 |
| |
| 摘 要： | 借助于牛顿向后插公式对文 1 ]的 NFI-AMS学习算法进行了改进 ,改进后的联想记忆系统的学习算法不但具有原来学习算法的收敛速度快、学习精度高等优点 ,而且还具备了 CMAC-AMS本身具有的局部泛化 (generalaization亦称推广 )能力以及对周围信息的收集能力大大增强等等 .数值模拟表明 ,这种改进的NFI-AMS在信号处理、模式识别及高精度的实时智能控制等领域具有很大的应用潜力
|
| 关 键 词： | 神经网络 联想记忆系统 函数逼近 牛顿向后插公式 |
| 修稿时间： | 2003年4月28日 |
Improvement of Learning Algorithm for the Associative Memory System |
| |
| XING Chun-feng,WANG Xin-feng.Improvement of Learning Algorithm for the Associative Memory System[J].Mathematics in Practice and Theory,2004,34(9):92-96. |
| |
| Authors: | XING Chun-feng WANG Xin-feng |
| |
| Abstract: | The Associative Memory System based on Newton′s Forward Interpolation formula(NFI-AMS) is improved by Newton's backward Interpolation formula. Improved learning algorithm posses not only the advantages over NFI-AMS in fast convergence rate and high precision of learning, but also the advantages over part generalaization of CMAC-AMS and strengthens collection information. Numerical simulations have shown that application areas of signal processing, pattern recognition, and controller implementation for high-precision real-time intelligent control. |
| |
| Keywords: | neural network associative memory system function approximation Newton′s backward Interpolation formula |
| 本文献已被 CNKI 万方数据 等数据库收录！ |
|
