Comparison of Methods to Produce Sample Points in Training ANN

Wang Shao-bo · Journal of Zhengzhou University of Technology · 2003

Based on the theory of the NN in the biology,the ANN is a complicated,massive,nonlinear dynamic system that simulates the biologic neural structureLots of sample points are required to train the ANNIn this paper, three methods are presented to produce the sample points adopted in training the ANNIt is testified that under the same condition even design is the best method to produce sample points,followed by orthogonal design,and the third is ransackingRansacking will be better if the number of sample points is largeWhen the function changes little with the variables,fewer factor levels can be chosen and orthogonal design is a good methodIts advantages stand out for the occasion of multiple variables and many factor levels.

Read the paper · More papers on PaperTik