具有双储层结构的动态误差补偿回声状态网络
A new echo state network with a double reservoir compensates for dynamic error
张昭昭 1朱应钦 2余文3
作者信息
- 1. 西安科技大学计算机科学与计算学院,陕西西安 710054
- 2. 西安科技大学计算机科学与计算学院,陕西西安 710054;墨西哥国立理工大学高级研究中心(CINVESTAV)自动化研究所,墨西哥城07360
- 3. 墨西哥国立理工大学高级研究中心(CINVESTAV)自动化研究所,墨西哥城07360
- 折叠
摘要
针对传统回声状态网络难以有效应对高阶非线性复杂模型问题,本文在理论分析的基础上提出了一种双储层结构的误差补偿回声状态网络,并设计了该网络的学习算法.该网络由计算层和补偿层构成,计算层主要承担拟合任务,补偿层则作为状态跟随器,实时补偿由于计算层对期望方差估计不足而导致的幅值偏差.对多阶振荡器和真实高阶非线性数据集的实验结果表明,本文所提网络结构较常规网络具有更高的稳定性和泛化性能,尤其对高阶非线性复杂模型的预测精度大幅度提升.
Abstract
The traditional echo state network is challenging to deal with the high-order nonlinear complex model effec-tively.We proposed an error trace reservoir computing and designed the optimal network algorithm.This new reservoir computing structure consists of a computing layer and a compensation layer.The computing layer mainly undertakes the fitting task,and the compensation layer acts as an error trace function.Because the computing layer always has an insuffi-cient variance estimation,it will lead to unstable neural network prediction.Thus,we proposed the compensation layer to trace neural network error in real-time.The numerical experiments on modeling the multiple superimposed oscillators and nonlinear data sets demonstrate that error trace reservoir structure has higher stability and generalization performance than the conventional network,especially in the high order nonlinear complex models.
关键词
回声状态网络/高阶非线性复杂模型/补偿回声状态网络/多阶振荡器Key words
echo state network/high order nonlinear complex model/compensating echo state network/multiple superimposed oscillator引用本文复制引用
基金项目
陕西省自然科学基础研究计划陕煤联合基金(2019JLZ-08)
陕西省自然科学基础研究计划(2020JM-522)
陕西省自然科学基础研究计划(2021JM-396)
出版年
2024
国家科技期刊平台
NSTL
万方数据