摘要
C_CWang的多值指数双向联想记忆模型 (MVeBAM)是一种高存储容量的联想神经网络 .本文在MVe BAM的基础上通过引入自相关项 (或内连接 )提出了一个新的具有内连接的多值指数双向联想记忆模型 ,推广了MVeBAM .通过定义简单的能量函数证明了其在同、异步方式下的稳定性 ,从而保证了所学模式对成为被推广的MVeBAM(EMVeBAM)的稳定点 .最后 。
C_C Wang's multi_valued exponential bidirectional associative memory model (MVeBAM) is a neural network with higher storage capacity. In this paper, based on the MVeBAM, we propose a new multi_valued exponential bidirectional associative memory model with intraconnection (EMVeBAM) by adding an auto_correlation term (or an intraconnection) to the exponents, extending the MVeBAM. The stability of the proposed model is proven in synchronous and asynchronous update modes with a defined energy function, which ensures that the learnt patterns become stable points of the EMeBAM. Finally, the computer simulation results verify that the EMVeBAM has higher storage capacity and better error_correcting capability than those of MVeBAM.
出处
《控制理论与应用》
EI
CAS
CSCD
北大核心
2002年第1期65-67,72,共4页
Control Theory & Applications
基金
国家自然科学基金 (6970 10 0 4)
教育部青年骨干教师资助计划资助项目
关键词
稳定性
联想神经网络
指数多值双向联想记忆模型
能量函数
multi_valued) bidirectional associative memory
exponent
neural networks
stability
intraconnection