Supervised contrastive representation learning with tree-structured parzen estimator Bayesian optimization for imbalanced tabular data

Shuting Tao, Peng Peng, Yunfei Li, Haiyue Sun, Qi Li, Hongwei Wang

Abstract: Highlights•We combine contrastive loss and Bayesian optimization for imbalanced tabular data.•Supervised contrastive loss solves the issue of lack of tabular data augmentation.•Tree-Structured Parzen Estimator is proven efficient in searching temperature.•The proposed method achieves remarkable success on 15 imbalanced public datasets.•Ablations confirm novel integration of contrastive loss and Bayesian optimization.