Feed-back neural networks with discrete weights

CORC > 厦门大学 > 物理技术－已发表论文

	Feed-back neural networks with discrete weights
	Wang, Lijuan ; Zhou, Qingguo ; Jin, Tao ; Zhao, Hong ; Zhao H(赵鸿)
刊名	http://dx.doi.org/10.1007/s00521-012-0867-8
	2013
关键词	ASSOCIATIVE MEMORY SYNAPSES SATISFIABILITY COUPLINGS CAPACITY DYNAMICS MATRICES MODELS
英文摘要	National Natural Science Foundation of China [60973137, 10775115, 10975115, 10925525]; Gansu Sci. & Tech. Support Program [1104GKCA049]; Fundamental Research Funds for the Central Universities [lzujbky-2010-89]; Google; We use the Monte Carlo Adaptation learning algorithm to design feed-back neural networks with discrete weights. The dynamic properties of these types of neural networks are investigated as a function of the states of weights. The numerical results of these networks show three phases: the "chaos phase," the "pure memory phase" and the "mixture phase" in the parameter space. The maximum storage ratio for the "pure memory phase" increases with the increasing of the states of the weights, which is favorable for practical applications.
语种	英语
出版者	SPRINGER
内容类型	期刊论文
源URL	[http://dspace.xmu.edu.cn/handle/2288/91910]
专题	物理技术－已发表论文
推荐引用方式 GB/T 7714	Wang, Lijuan,Zhou, Qingguo,Jin, Tao,et al. Feed-back neural networks with discrete weights[J]. http://dx.doi.org/10.1007/s00521-012-0867-8,2013.
APA	Wang, Lijuan,Zhou, Qingguo,Jin, Tao,Zhao, Hong,&赵鸿.(2013).Feed-back neural networks with discrete weights.http://dx.doi.org/10.1007/s00521-012-0867-8.
MLA	Wang, Lijuan,et al."Feed-back neural networks with discrete weights".http://dx.doi.org/10.1007/s00521-012-0867-8 (2013).