Copula for Instance-wise Feature Selection and Ranking

TL;DR

This paper introduces a novel feature selection method using Gaussian copula to model feature dependencies, improving the selection process for neural network tasks with better interpretability and performance.

Contribution

It integrates Gaussian copula into instance-wise feature selection, effectively capturing feature dependencies without extra modifications.

Findings

Captures meaningful feature correlations in synthetic and real datasets.

Enhances interpretability of feature selection results.

Improves task performance in neural networks.

Abstract

Instance-wise feature selection and ranking methods can achieve a good selection of task-friendly features for each sample in the context of neural networks. However, existing approaches that assume feature subsets to be independent are imperfect when considering the dependency between features. To address this limitation, we propose to incorporate the Gaussian copula, a powerful mathematical technique for capturing correlations between variables, into the current feature selection framework with no additional changes needed. Experimental results on both synthetic and real datasets, in terms of performance comparison and interpretability, demonstrate that our method is capable of capturing meaningful correlations.