XTab: Cross-table Pretraining for Tabular Transformers
This addresses the challenge of leveraging information across multiple tables for tabular data, improving generalizability and performance in tasks like regression and classification.
The paper tackles the problem of limited generalization in tabular self-supervised learning by introducing XTab, a cross-table pretraining framework for tabular transformers, which boosts performance on 84 tasks and achieves superior results compared to state-of-the-art models.
The success of self-supervised learning in computer vision and natural language processing has motivated pretraining methods on tabular data. However, most existing tabular self-supervised learning models fail to leverage information across multiple data tables and cannot generalize to new tables. In this work, we introduce XTab, a framework for cross-table pretraining of tabular transformers on datasets from various domains. We address the challenge of inconsistent column types and quantities among tables by utilizing independent featurizers and using federated learning to pretrain the shared component. Tested on 84 tabular prediction tasks from the OpenML-AutoML Benchmark (AMLB), we show that (1) XTab consistently boosts the generalizability, learning speed, and performance of multiple tabular transformers, (2) by pretraining FT-Transformer via XTab, we achieve superior performance than other state-of-the-art tabular deep learning models on various tasks such as regression, binary, and multiclass classification.