Nonlinear feature selection using sparsity-promoted centroid-encoder

The contribution of our work is two-fold. First, we propose a novel feature selection technique, sparsity-promoted centroid-encoder (SCE). The model uses the nonlinear mapping of artificial neural networks to reconstruct a sample as its class centroid and, at the same time, apply a ℓ₁-penalty to the weights of a sparsity promoting layer, placed between the input and first hidden layer, to select discriminative features from input data. Using the proposed method, we designed a feature selection framework that first ranks each feature and then, compiles the optimal set using validation samples. The second part of our study investigates the role of stochastic optimization, such as Adam, in minimizing ℓ₁-norm. The empirical analysis shows that the hyper-parameters of Adam (mini-batch size, learning rate, etc.) play a crucial role in promoting feature sparsity by SCE. We apply our technique to numerous real-world data sets and find that it significantly outperforms other state-of-the-art methods, including LassoNet, stochastic gates (STG), feature selection networks (FsNet), supervised concrete autoencoder (CAE), deep feature selection (DFS), and random forest (RF).