RNN与提前终止迭代在金属岩石分类中的比较

Comparative Analysis Between Regularized Neural Network Algorithm and Early Stopping Iteration Algorithm on Metal and Rock Classification

神经网络算法是一种非常经典的分类算法,然而神经网络的一个不足之处就是容易陷入过拟合。针对这种不足,正则化神经网路算法(RNM)与提前终止迭代算法被提了出来。为了进一步研究这两种算法性能的差异,本文在包括金属岩石分类数据集(sonar数据集)在内的20个UCI标准数据集上对这两种方法进行了性能测试。实验显示在分类准确率上正则化神经网路算法要更优秀一些,但是在分类速度上提前终止迭代算法更占优势。

The neural network algorithm is a very classic classification algorithm However,over- fitting is easy to arisen for neural network algorithm for this shortfall,regularized neural network algorithm and early termination ofthe iterative method was proposed inorder to further study the differences of performance between these two algorithms,in this papel we use 20 UCI standard data sets including metal and rock classification dataset （sonar dataset）to test the performance ofthe two methods the experiments show that the regularization neural network algorithm exhibits superiority over the early stopping iteration algorithm at classification accuracy,but the early stopping iteration algorithm is much better at the classification speed.