An Augmentation-Aware Theory for Self-Supervised Contrastive Learning
This work addresses a theoretical gap for researchers in self-supervised learning, though it is incremental as it builds on existing frameworks.
The paper tackles the under-explored role of specific data augmentation types in self-supervised contrastive learning by proposing an augmentation-aware error bound that links supervised risk to unsupervised risk and augmentation trade-offs, with experiments validating the theoretical results.
Self-supervised contrastive learning has emerged as a powerful tool in machine learning and computer vision to learn meaningful representations from unlabeled data. Meanwhile, its empirical success has encouraged many theoretical studies to reveal the learning mechanisms. However, in the existing theoretical research, the role of data augmentation is still under-exploited, especially the effects of specific augmentation types. To fill in the blank, we for the first time propose an augmentation-aware error bound for self-supervised contrastive learning, showing that the supervised risk is bounded not only by the unsupervised risk, but also explicitly by a trade-off induced by data augmentation. Then, under a novel semantic label assumption, we discuss how certain augmentation methods affect the error bound. Lastly, we conduct both pixel- and representation-level experiments to verify our proposed theoretical results.