Feature Augmentation for Self-supervised Contrastive Learning: A Closer Look
Abstract
Self-supervised contrastive learning heavily relies on the view variance brought by data augmentation, so that it can learn a view-invariant pre-trained representation. Beyond increasing the view variance for contrast, this work focuses on improving the diversity of training data, to improve the generalization and robustness of the pre-trained models. To this end, we propose a unified framework to conduct data augmentation in the feature space, known as feature augmentation. This strategy is domain-agnostic, which augments similar features to the original ones and thus improves the data diversity. We perform a systematic investigation of various feature augmentation architectures, the gradient-flow skill, and the relationship between feature augmentation and traditional data augmentation. Our study reveals some practical principles for feature augmentation in self-contrastive learning. By integrating feature augmentation on the instance discrimination or the instance similarity paradigm, we consistently improve the performance of pre-trained feature learning and gain better generalization over the downstream image classification and object detection task.
Cite
@article{arxiv.2410.12396,
title = {Feature Augmentation for Self-supervised Contrastive Learning: A Closer Look},
author = {Yong Zhang and Rui Zhu and Shifeng Zhang and Xu Zhou and Shifeng Chen and Xiaofan Chen},
journal= {arXiv preprint arXiv:2410.12396},
year = {2024}
}
Comments
IJCNN 2024