Safe Semi-Supervised Contrastive Learning Using In-Distribution Data as Positive Examples

TL;DR

This paper introduces a self-supervised contrastive learning method for semi-supervised classification that effectively utilizes in-distribution data as positive examples, especially under class distribution mismatch scenarios with out-of-distribution data.

Contribution

It proposes a novel contrastive loss with a coefficient schedule to leverage in-distribution data as positive examples, improving performance in semi-supervised learning with distribution mismatch.

Findings

Self-supervised contrastive learning significantly improves classification accuracy.

Aggregating in-distribution examples yields better representations.

The method performs well across various datasets and mismatch ratios.

Abstract

Semi-supervised learning methods have shown promising results in solving many practical problems when only a few labels are available. The existing methods assume that the class distributions of labeled and unlabeled data are equal; however, their performances are significantly degraded in class distribution mismatch scenarios where out-of-distribution (OOD) data exist in the unlabeled data. Previous safe semi-supervised learning studies have addressed this problem by making OOD data less likely to affect training based on labeled data. However, even if the studies effectively filter out the unnecessary OOD data, they can lose the basic information that all data share regardless of class. To this end, we propose to apply a self-supervised contrastive learning approach to fully exploit a large amount of unlabeled data. We also propose a contrastive loss function with coefficient schedule to aggregate as an anchor the labeled negative examples of the same class into positive examples. To evaluate the performance of the proposed method, we conduct experiments on image classification datasets - CIFAR-10, CIFAR-100, Tiny ImageNet, and CIFAR-100+Tiny ImageNet - under various mismatch ratios. The results show that self-supervised contrastive learning significantly improves classification accuracy. Moreover, aggregating the in-distribution examples produces better representation and consequently further improves classification accuracy.