Improving Multi-Label Contrastive Learning by Leveraging Label Distribution
This addresses a key bottleneck in multi-label learning for applications like image and vector data analysis, though it appears incremental as it builds on existing contrastive learning frameworks.
The paper tackles the challenge of selecting positive/negative samples and utilizing label information in multi-label contrastive learning by proposing a method that leverages label distribution, requiring only label intersection checks and recovering distributions via RBF or contrastive loss. It outperforms state-of-the-art methods on nine datasets across six evaluation metrics.
In multi-label learning, leveraging contrastive learning to learn better representations faces a key challenge: selecting positive and negative samples and effectively utilizing label information. Previous studies selected positive and negative samples based on the overlap between labels and used them for label-wise loss balancing. However, these methods suffer from a complex selection process and fail to account for the varying importance of different labels. To address these problems, we propose a novel method that improves multi-label contrastive learning through label distribution. Specifically, when selecting positive and negative samples, we only need to consider whether there is an intersection between labels. To model the relationships between labels, we introduce two methods to recover label distributions from logical labels, based on Radial Basis Function (RBF) and contrastive loss, respectively. We evaluate our method on nine widely used multi-label datasets, including image and vector datasets. The results demonstrate that our method outperforms state-of-the-art methods in six evaluation metrics.