SimMatch: Semi-supervised Learning with Similarity Matching

TL;DR

SimMatch introduces a semi-supervised learning framework that leverages both semantic and instance similarities through consistency regularization, a labeled memory buffer, and similarity transformation operations, leading to state-of-the-art results on benchmark datasets.

Contribution

The paper proposes a novel semi-supervised learning framework, SimMatch, that jointly considers semantic and instance similarities with new operations for improved label propagation.

Findings

Achieves 67.2% Top-1 accuracy with 1% labeled data on ImageNet

Outperforms previous semi-supervised learning methods

Demonstrates effectiveness across multiple datasets and settings

Abstract

Learning with few labeled data has been a longstanding problem in the computer vision and machine learning research community. In this paper, we introduced a new semi-supervised learning framework, SimMatch, which simultaneously considers semantic similarity and instance similarity. In SimMatch, the consistency regularization will be applied on both semantic-level and instance-level. The different augmented views of the same instance are encouraged to have the same class prediction and similar similarity relationship respected to other instances. Next, we instantiated a labeled memory buffer to fully leverage the ground truth labels on instance-level and bridge the gaps between the semantic and instance similarities. Finally, we proposed the \textit{unfolding} and \textit{aggregation} operation which allows these two similarities be isomorphically transformed with each other. In this way, the semantic and instance pseudo-labels can be mutually propagated to generate more high-quality and reliable matching targets. Extensive experimental results demonstrate that SimMatch improves the performance of semi-supervised learning tasks across different benchmark datasets and different settings. Notably, with 400 epochs of training, SimMatch achieves 67.2\%, and 74.4\% Top-1 Accuracy with 1\% and 10\% labeled examples on ImageNet, which significantly outperforms the baseline methods and is better than previous semi-supervised learning frameworks. Code and pre-trained models are available at https://github.com/KyleZheng1997/simmatch.