Contrastive Learning with Negative Sampling Correction
This addresses a specific bottleneck in contrastive learning for representation learning, offering an incremental improvement over existing methods.
The paper tackles the problem of negative sampling bias in contrastive learning, where generated negative samples are often polluted by positive samples, leading to performance degradation. It proposes Positive-Unlabeled Contrastive Learning (PUCL), which corrects this bias by treating negative samples as unlabeled and using positive sample information, resulting in state-of-the-art performance on various image and graph classification tasks.
As one of the most effective self-supervised representation learning methods, contrastive learning (CL) relies on multiple negative pairs to contrast against each positive pair. In the standard practice of contrastive learning, data augmentation methods are utilized to generate both positive and negative pairs. While existing works have been focusing on improving the positive sampling, the negative sampling process is often overlooked. In fact, the generated negative samples are often polluted by positive samples, which leads to a biased loss and performance degradation. To correct the negative sampling bias, we propose a novel contrastive learning method named Positive-Unlabeled Contrastive Learning (PUCL). PUCL treats the generated negative samples as unlabeled samples and uses information from positive samples to correct bias in contrastive loss. We prove that the corrected loss used in PUCL only incurs a negligible bias compared to the unbiased contrastive loss. PUCL can be applied to general contrastive learning problems and outperforms state-of-the-art methods on various image and graph classification tasks. The code of PUCL is in the supplementary file.