TFWT: Tabular Feature Weighting with Transformer

TL;DR

This paper introduces TFWT, a transformer-based feature weighting method for tabular data that captures complex feature dependencies and uses reinforcement learning for fine-tuning, improving performance on real-world datasets.

Contribution

The paper proposes a novel transformer-based feature weighting approach with reinforcement learning for tabular data, addressing limitations of existing methods that assume equal feature importance.

Findings

TFWT outperforms existing feature weighting methods on multiple datasets.

Transformer-based weighting captures complex feature dependencies effectively.

Reinforcement learning enhances the fine-tuning of feature weights.

Abstract

In this paper, we propose a novel feature weighting method to address the limitation of existing feature processing methods for tabular data. Typically the existing methods assume equal importance across all samples and features in one dataset. This simplified processing methods overlook the unique contributions of each feature, and thus may miss important feature information. As a result, it leads to suboptimal performance in complex datasets with rich features. To address this problem, we introduce Tabular Feature Weighting with Transformer, a novel feature weighting approach for tabular data. Our method adopts Transformer to capture complex feature dependencies and contextually assign appropriate weights to discrete and continuous features. Besides, we employ a reinforcement learning strategy to further fine-tune the weighting process. Our extensive experimental results across various real-world datasets and diverse downstream tasks show the effectiveness of TFWT and highlight the potential for enhancing feature weighting in tabular data analysis.