Geodesic Flow Kernels for Semi-Supervised Learning on Mixed-Variable Tabular Dataset

TL;DR

This paper introduces GFTab, a semi-supervised learning framework for mixed-variable tabular data that uses geodesic flow kernels and variable-specific corruption to improve performance with limited labels.

Contribution

GFTab is the first to combine geodesic flow kernels, variable-specific corruption, and tree-based embeddings for semi-supervised learning on mixed-variable tabular datasets.

Findings

GFTab outperforms existing models on 21 diverse datasets.

GFTab is especially effective with limited labeled data.

The framework captures geometric relationships in corrupted inputs.

Abstract

Tabular data poses unique challenges due to its heterogeneous nature, combining both continuous and categorical variables. Existing approaches often struggle to effectively capture the underlying structure and relationships within such data. We propose GFTab (Geodesic Flow Kernels for Semi- Supervised Learning on Mixed-Variable Tabular Dataset), a semi-supervised framework specifically designed for tabular datasets. GFTab incorporates three key innovations: 1) Variable-specific corruption methods tailored to the distinct properties of continuous and categorical variables, 2) A Geodesic flow kernel based similarity measure to capture geometric changes between corrupted inputs, and 3) Tree-based embedding to leverage hierarchical relationships from available labeled data. To rigorously evaluate GFTab, we curate a comprehensive set of 21 tabular datasets spanning various domains, sizes, and variable compositions. Our experimental results show that GFTab outperforms existing ML/DL models across many of these datasets, particularly in settings with limited labeled data.