LassoLayer: Nonlinear Feature Selection by Switching One-to-one Links

TL;DR

LassoLayer introduces a nonlinear feature selection method using a one-to-one link structure trained by L1 optimization, integrated into neural networks to improve feature selection in complex tasks.

Contribution

The paper presents LassoLayer and LassoMLP, enabling nonlinear feature selection within neural networks, extending Lasso's capabilities beyond linear models.

Findings

LassoMLP outperforms state-of-the-art methods on MNIST.

LassoLayer effectively drops unnecessary units for prediction.

Applicable to various network structures for feature selection.

Abstract

Along with the desire to address more complex problems, feature selection methods have gained in importance. Feature selection methods can be classified into wrapper method, filter method, and embedded method. Being a powerful embedded feature selection method, Lasso has attracted the attention of many researchers. However, as a linear approach, the applicability of Lasso has been limited. In this work, we propose LassoLayer that is one-to-one connected and trained by L1 optimization, which work to drop out unnecessary units for prediction. For nonlinear feature selections, we build LassoMLP: the network equipped with LassoLayer as its first layer. Because we can insert LassoLayer in any network structure, it can harness the strength of neural network suitable for tasks where feature selection is needed. We evaluate LassoMLP in feature selection with regression and classification tasks. LassoMLP receives features including considerable numbers of noisy factors that is harmful for overfitting. In the experiments using MNIST dataset, we confirm that LassoMLP outperforms the state-of-the-art method.