Tabular ML

ML: Machine Learning

Overview

Tabular machine learning studies prediction, discovery, and explanation from structured data with numerical, categorical, text, and missing values. Common approaches include gradient-boosted decision trees, neural networks with feature embeddings and attention, pretrained or in-context models, and automated feature engineering. Benchmarks compare these methods on classification, regression, anomaly detection, and specialized tasks, but no model family consistently leads across dataset sizes, feature types, and settings. A central challenge is making results generalize beyond clean IID benchmarks while balancing accuracy with robustness, interpretability, computational cost, and careful evaluation.

Latest papers

CardsList

125 papers9 papers in the last four weeks, down 47% on the four weeks before. 0.1% of all new papers.

  1. Hippasus: Accurate and Cost-Efficient Feature Augmentation over Relational Tables

    Feb 2, 2026Serafeim Papadias, Kostas Patroumpas, Dimitrios SkoutasAutomated Feature EngineeringTabular ML

  2. Integrated Multivariate Segmentation Tree for Heterogeneous Credit Data Analysis in Small- and Medium-Sized Enterprises

    Aug 30, 2025Lu Han, Xiuying WangFinancial ServicesInterpretable ML

  3. Scalable Source Selection for Machine Learning Tasks

    Jul 29, 2025Ambarish Singh, Romila PradhanTraining Data SelectionTabular ML

  4. Predicting Estimated Times of Restoration for Electrical Outages Using Longitudinal Tabular Transformers

    May 1, 2025Bogireddy Sai Prasanna Teja, Valliappan Muthukaruppan, Carls BenjaminEvent Time PredictionTransformer

  5. Effector: A Python Package for Global and Regional Feature Effects

    Apr 3, 2024Vasilis Gkolemis, Christos Diou, Dimitris Kyriakopoulos +10Tabular ML