摘要
提出并用软件实现了一种时变容错域的感知联想记忆模型.该模型有以下特点:(1)模拟了大脑联想记忆的容错域随时间变化的特点.模型可以根据记忆样本的重要性,为记忆样本在不同时间设计适当的容错域;(2)实现了n维到m维空间的无穷值模式非线性联想,而且该模型的样本容错域充满整个实空间Rn,模型没有伪吸引子;(3)联想记忆速度快.
A time varying fault-tolerant field perceptive associative memory model is proposed and realized by software. One of the characters of this model is that it simulates the time-dependent changeability character of the fault-tolerant field of human brain＇s associative memory. This model can design proper fault-tolerant field for memory sample at different time according to the essentiality of memory sample. The second character is that it realizes the nonlinear association of infinite value pattern from n dimension space to m dimension space. And the fault-tolerant field of the memory samples is full of the whole real space R^n and the associative memory model has no spurious attractors. The third character is that the speed of associative memory about the model is faster.
出处
《计算机学报》
EI
CSCD
北大核心
2006年第3期431-440,共10页
Chinese Journal of Computers
基金
国家自然科学基金(60375014)
山东省自然科学基金(Y2002G16)
中国博士后科学基金资助.
关键词
联想记忆
神经网络
时变容错域
人工生命
人工脑
associative memory
neural network
time varying fault-tolerant field
artificial life
artificial brain
