Meta Pseudo Labels

Hieu Pham, Zihang Dai, Qizhe Xie, Minh-Thang Luong, Quoc V. Le
3/23/2020
cs.LGstat.ML

Official code: google-research/google-research (39,000 stars, Apache-2.0, last commit 2026-04-30).

Abstract

We present Meta Pseudo Labels, a semi-supervised learning method that achieves a new state-of-the-art top-1 accuracy of 90.2% on ImageNet, which is 1.6% better than the existing state-of-the-art. Like Pseudo Labels, Meta Pseudo Labels has a teacher network to generate pseudo labels on unlabeled data to teach a student network. However, unlike Pseudo Labels where the teacher is fixed, the teacher in Meta Pseudo Labels is constantly adapted by the feedback of the student's performance on the labeled dataset. As a result, the teacher generates better pseudo labels to teach the student. Our code will be available at https://github.com/google-research/google-research/tree/master/meta_pseudo_labels.

View on arXivView PDF

Code Implementations(1)

google-research/google-researchOfficial100%
Apache-2.0

Google Research

39,0008,488C, ROct 4, 20185 months agoApache-2.0
aimachine-learningresearch

Similar Papers

Exploring Beyond Logits: Hierarchical Dynamic Labeling Based on Embeddings for Semi-Supervised Classification

87% match2024

Collaboration of Teachers for Semi-supervised Object Detection

87% match2024

Hierarchical Supervision and Shuffle Data Augmentation for 3D Semi-Supervised Object Detection

86% match1 repo712023

CoMatch: Semi-supervised Learning with Contrastive Graph Regularization

85% match1 repo1332020

Learning Adaptive Pseudo-Label Selection for Semi-Supervised 3D Object Detection

85% match2025

Synergy-Guided Regional Supervision of Pseudo Labels for Semi-Supervised Medical Image Segmentation

85% match2024

Cite this paper

@article{pham2020meta,
  title  = {Meta Pseudo Labels},
  author = {Hieu Pham and Zihang Dai and Qizhe Xie and Minh-Thang Luong and Quoc V. Le},
  year   = {2020},
  eprint = {2003.10580},
  archivePrefix = {arXiv},
  url    = {https://arxiv.org/abs/2003.10580}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2003.10580.svg)](https://codeofpaper.com/paper/2003.10580)

Discussion