Strong classifier design based on BP_Adaboost--fin
2016-08-23
3 0 0
no vote
Other
Earn points
BP neural network is a reverse pass and multi-level mapping function of correcting errors, it adopted after learning of the unknown system of input and output parameters, we can express the associative memory system. But because of BP network based on the gradient method is derived from requiring objective function is continuously differentiable, skilled in the evolutionary process of learning is slow and easily get into local Optima, can't find globally optimal solutions. And because of the BP network weights and thresholds is a random value in the select, each has a different initial value, resulting in every practice learning prediction results are different. Adaboost algorithm can improve the accuracy of classification of any given weak classifiers in many machine learning problems have been successfully applied. In order to improve the classification accuracy of BP network, overcoming the limitations of the weights of BP network initialization and subjective factors of training sa
matlab
分类
财务
基于
建模
预警
设计公司
BPAdaboost
Related Source Codes
GMSK Linear Receiver
0
0
no vote
NSGA-II algorithm
0
0
no vote
NSGA-III multi-objective optimization algorithm
0
0
no vote
Compressed sensing example
0
0
no vote
CFAR detector example
0
0
no vote
No comment