通过特征空间增强提升自监督对比学习的泛化能力
Feature Augmentation for Self-supervised Contrastive Learning: A Closer Look
- 在特征空间进行数据增强，提升训练数据多样性
- 结合实例判别与相似性任务，显著提升下游任务性能
- 方法通用性强，适用于多种图像识别场景
自监督对比学习高度依赖数据增强带来的视角差异，以学习视角不变的预训练表征。本文不仅关注增加对比视角的方差，更聚焦于提升训练数据的多样性，以增强预训练模型的泛化性和鲁棒性。为此，提出一种统一的特征空间增强框架——特征增强，该方法不依赖具体领域，通过生成与原特征相似的新特征来提升数据多样性。系统研究了多种特征增强架构、梯度流动技巧，以及特征增强与传统数据增强之间的关系。研究揭示了特征增强在自对比学习中的实用原则。将特征增强应用于实例判别或实例相似性范式，均能持续提升预训练特征学习性能，在下游图像分类和目标检测任务中实现更好泛化。
原文摘要 · Abstract (English)
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.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。