Generative Table Pre-training Empowers Models for Tabular Prediction
This work addresses the problem of enhancing tabular prediction for data scientists by offering a versatile pre-training method, though it appears incremental as it builds on existing table pre-training concepts.
The authors tackled the challenge of using table pre-training to improve tabular prediction by proposing TapTap, which generates synthetic tables and outperforms 16 baselines across 12 datasets, with models using synthetic data competing with those using original data on half the datasets.
Recently, the topic of table pre-training has attracted considerable research interest. However, how to employ table pre-training to boost the performance of tabular prediction remains an open challenge. In this paper, we propose TapTap, the first attempt that leverages table pre-training to empower models for tabular prediction. After pre-training on a large corpus of real-world tabular data, TapTap can generate high-quality synthetic tables to support various applications on tabular data, including privacy protection, low resource regime, missing value imputation, and imbalanced classification. Extensive experiments on 12 datasets demonstrate that TapTap outperforms a total of 16 baselines in different scenarios. Meanwhile, it can be easily combined with various backbone models, including LightGBM, Multilayer Perceptron (MLP) and Transformer. Moreover, with the aid of table pre-training, models trained using synthetic data generated by TapTap can even compete with models using the original dataset on half of the experimental datasets, marking a milestone in the development of synthetic tabular data generation. The codes are available at https://github.com/ZhangTP1996/TapTap.