Unsupervised Representation Learning by InvariancePropagation

TL;DR

This paper introduces Invariance Propagation, a novel unsupervised learning method that emphasizes learning category-level invariance by recursively discovering semantically similar samples, leading to improved performance on various image classification tasks.

Contribution

The paper proposes a new approach called Invariance Propagation that focuses on category-level invariance and uses a hard sampling strategy to enhance representation learning.

Findings

Achieves 71.3% top-1 accuracy on ImageNet linear classification.

Surpasses previous methods on downstream tasks like Places205 and Pascal VOC.

Effective in transfer learning on small datasets.

Abstract

Unsupervised learning methods based on contrastive learning have drawn increasing attention and achieved promising results. Most of them aim to learn representations invariant to instance-level variations, which are provided by different views of the same instance. In this paper, we propose Invariance Propagation to focus on learning representations invariant to category-level variations, which are provided by different instances from the same category. Our method recursively discovers semantically consistent samples residing in the same high-density regions in representation space. We demonstrate a hard sampling strategy to concentrate on maximizing the agreement between the anchor sample and its hard positive samples, which provide more intra-class variations to help capture more abstract invariance. As a result, with a ResNet-50 as the backbone, our method achieves 71.3% top-1 accuracy on ImageNet linear classification and 78.2% top-5 accuracy fine-tuning on only 1% labels, surpassing previous results. We also achieve state-of-the-art performance on other downstream tasks, including linear classification on Places205 and Pascal VOC, and transfer learning on small scale datasets.