| 引用本文： |
-
周永权,刘宣会.基于Hensel构造的回归神经网络符号计算模型及算法[J].广西科学,2003,10(3):176-178,182. [点击复制]
- Zhou Yongquan,Liu Xuanhui.Recurrent Neural Networks Symbol Computation Model and Algorithm Based on Hensel Construction[J].Guangxi Sciences,2003,10(3):176-178,182. [点击复制]
|
|
| 摘要: |
| 将传统意义下Hensel构造提升的方法与回归神经网络模型有机地结合起来,提出一种基于Hensel构造方法的回归神经网络近似代数符号计算新模型和PFRNN网络算法.该模型不但具有回归神经网络的特点,而且具有Hensel构造提升的思想,给人们研究代数符号计算与近似代数符号计算提供一种可视化手段.通过多元多项式近似因式分解算例分析可以看出,新模型刻划出在符号计算意义下精确计算与近似计算的本质与联系. |
| 关键词: 回归神经网络 Hensel构造方法 近似分解 代数符号计算 |
| DOI： |
| 投稿时间：2002-11-18 |
| 基金项目:广西自然科学基金(0141034)及广西高校百名中青年学科带头人的资助。 |
|
| Recurrent Neural Networks Symbol Computation Model and Algorithm Based on Hensel Construction |
|
Zhou Yongquan, Liu Xuanhui
|
| (Dept. of Math. & Comp. Sci., Guangxi Univ. for Nationalities, 80 Daxuelu, Nanning, Guangxi, 530006, China) |
| Abstract: |
| Under the traditional Hensel construction method and recurrent neural networks model,a new recurrent neural networks symbol algebra symbol computation model and PFRNN Algorithm based on Hensel construction is proposed.It has the characteristics of traditional RNN and the capability of function approximation,and may offer a kind of visual means for studying algebra symbol calculation and approximate algebra symbol calculation.Through multivariate polynomials approximate factorization,the essences and relationships between approximate calculation and accurate calculation are explained. |
| Key words: recurrent neural networks Hensel construction method approximate factorization algebra symbol computation |
