Local Contrastive Feature learning for Tabular Data
This addresses feature learning for tabular data, which is incremental as it adapts contrastive methods from other domains.
The paper tackles the problem of learning local patterns from tabular data without labels by proposing a local contrastive feature learning (LoCL) framework that uses feature correlations to create subsets and convolutional learning with contrastive and reconstruction losses. Experiments on public datasets show it outperforms state-of-the-art baselines.
Contrastive self-supervised learning has been successfully used in many domains, such as images, texts, graphs, etc., to learn features without requiring label information. In this paper, we propose a new local contrastive feature learning (LoCL) framework, and our theme is to learn local patterns/features from tabular data. In order to create a niche for local learning, we use feature correlations to create a maximum-spanning tree, and break the tree into feature subsets, with strongly correlated features being assigned next to each other. Convolutional learning of the features is used to learn latent feature space, regulated by contrastive and reconstruction losses. Experiments on public tabular datasets show the effectiveness of the proposed method versus state-of-the-art baseline methods.