摘要
云模型是定性定量间的不确定转换模型 ,它将概念的模糊性和随机性集成在一起。文中提出一种利用云模型来有效避免BP算法陷入局部极小的方法 ,该方法通过基于云模型和输入参数区间划分的学习因子自适应调整算法来实现。该算法在复杂非线性分类 (阴阳图 )
Clouds model is one for transformation between qualitative and quantitative knowledge,whose representation reflects fuzziness and randomness contained in linguistic concepts.In this paper,a clouds model is applied to BP algorithm improvement ,which avoids BP ANN trapping to local least value.This learning rate self-tunning method is based on clouds model and input parameter domain partition.Finally,this algorithm is simulated under complex non-linear classfication circumstances.
出处
《计算机仿真》
CSCD
2002年第3期123-126,共4页
Computer Simulation
