ReConTab: Regularized Contrastive Representation Learning for Tabular Data
This addresses the time-consuming and expertise-dependent feature engineering bottleneck in tabular data analysis, though it appears incremental as it builds on existing contrastive learning and autoencoder techniques.
The authors tackled the problem of manual feature engineering for tabular data by introducing ReConTab, a deep representation learning framework using regularized contrastive learning, which achieved substantial and robust performance improvements on real-world datasets and enhanced traditional methods like XGBoost and Random Forest.
Representation learning stands as one of the critical machine learning techniques across various domains. Through the acquisition of high-quality features, pre-trained embeddings significantly reduce input space redundancy, benefiting downstream pattern recognition tasks such as classification, regression, or detection. Nonetheless, in the domain of tabular data, feature engineering and selection still heavily rely on manual intervention, leading to time-consuming processes and necessitating domain expertise. In response to this challenge, we introduce ReConTab, a deep automatic representation learning framework with regularized contrastive learning. Agnostic to any type of modeling task, ReConTab constructs an asymmetric autoencoder based on the same raw features from model inputs, producing low-dimensional representative embeddings. Specifically, regularization techniques are applied for raw feature selection. Meanwhile, ReConTab leverages contrastive learning to distill the most pertinent information for downstream tasks. Experiments conducted on extensive real-world datasets substantiate the framework's capacity to yield substantial and robust performance improvements. Furthermore, we empirically demonstrate that pre-trained embeddings can seamlessly integrate as easily adaptable features, enhancing the performance of various traditional methods such as XGBoost and Random Forest.