摘要
研究了混沌时间序列预测问题.提出了一种由五元生长因子组调控的类皮层神经网络模型,即多簇回响状态网络模型(MCESN).研究表明该生长因子组能够有效决定模型的拓扑性质;同时具备小世界和无标度等复杂网络特征的MCESN能够获得较优的预测结果.通过Monte Carlo仿真实验表明,该模型不仅训练算法简单,而且与常规回响状态网络比较,预测结果的精度更高、标准差更小.
The chaotic time series prediction problem is considered. A novel type of cortex-like neural network model, i.e. multiclusters echo state network model （MCESN）, regulated by a group of five growth-factors, is proposed. It is shown that characters of MCESN＇ topology can be effectively determined by the growth-factors group; and that it is the MCESN possessing both smallworld and scale-free properties of complex network that corresponds to the better prediction performance. In addition, Monte Carlo simulation experiments show that MCESN not only can be trained by easy algorithm, but also can achieve higher accuracy and less standard deviation prediction results than classical echo state networks.
出处
《物理学报》
SCIE
EI
CAS
CSCD
北大核心
2009年第7期5057-5064,共8页
Acta Physica Sinica
基金
高等学校博士学科点专项科研基金(批准号:20050698032)
国家重点基础研究发展计划(973)项目(批准号:2007CB311006)
国家高技术研究发展计划(863)(批准号:2006AA04Z222)
国家自然科学基金(批准号:60875043)资助的课题~~
关键词
混沌时间序列预测
回响状态网络
复杂网络
Ω复杂性
chaotic time series prediction, echo state networks, complex networks, Omega complexity
