基于ℓ_2,1范数和神经网络的非线性特征选择方法

In joint ℓ_2,1-norm based feature selection methods, predicted attributes shared parallel sparsity coefficients and outliers could be removed. However, previous optimizing ℓ_2,1-norm solutions were usually based on linear solutions which could not discover the non-linear relationship between features. In this paper, we proposed a nonlinear solution for ℓ_2,1-norm based feature selection using the framework of neural network. With combination of ℓ_2,1-norm and neural network, on the one hand, we can make use of the nonlinearity of neural network to optimize ℓ_2,1-norm based problem. On the other hand, the ℓ_2,1-norm can help neural network to achieve feature selection. Finally, the proposed method is compared with famous feature selection methods on eight datasets. Experimental results empirically show the superiority of our method.