Pith. sign in

REVIEW 6 cited by

Big Self-Supervised Models are Strong Semi-Supervised Learners

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2006.10029 v2 pith:OG2UAXBF submitted 2020-06-17 cs.LG cs.CVstat.ML

classification cs.LGcs.CVstat.ML
keywords unlabeledexamplesfine-tuninglabelslearningsemi-supervisedaccuracydata
verification ladder T0 review T1 audit T2 compute T3 formal

Signed reviews

No signed human review yet.

0 comments
abstract

One paradigm for learning from few labeled examples while making best use of a large amount of unlabeled data is unsupervised pretraining followed by supervised fine-tuning. Although this paradigm uses unlabeled data in a task-agnostic way, in contrast to common approaches to semi-supervised learning for computer vision, we show that it is surprisingly effective for semi-supervised learning on ImageNet. A key ingredient of our approach is the use of big (deep and wide) networks during pretraining and fine-tuning. We find that, the fewer the labels, the more this approach (task-agnostic use of unlabeled data) benefits from a bigger network. After fine-tuning, the big network can be further improved and distilled into a much smaller one with little loss in classification accuracy by using the unlabeled examples for a second time, but in a task-specific way. The proposed semi-supervised learning algorithm can be summarized in three steps: unsupervised pretraining of a big ResNet model using SimCLRv2, supervised fine-tuning on a few labeled examples, and distillation with unlabeled examples for refining and transferring the task-specific knowledge. This procedure achieves 73.9% ImageNet top-1 accuracy with just 1% of the labels ($\le$13 labeled images per class) using ResNet-50, a $10\times$ improvement in label efficiency over the previous state-of-the-art. With 10% of labels, ResNet-50 trained with our method achieves 77.5% top-1 accuracy, outperforming standard supervised training with all of the labels.

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 6 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. Visual Pre-Training on Unlabeled Images using Reinforcement Learning

    cs.LG 2025-06 conditional novelty 6.0 of 10

    Casting image-crop consistency as temporal-difference value learning improves visual representations on unlabeled web, scene, and video data.

  2. Examining the Efficacy of Graph Neural Network Message-Passing in Regression Contexts

    cs.LG 2026-07 conditional novelty 5.0 of 10

    Across four NAS/DNN-predictor regression benchmarks, GEN (deep graph convolution) achieves the best average rank over 11 GNN message-passing layers, though attention GATv2 wins on the largest graphs.

  3. Long-Tailed Object Detection Pre-training: Dynamic Rebalancing Contrastive Learning with Dual Reconstruction

    cs.CV 2024-11 conditional novelty 5.0 of 10

    A detection pre-training framework that dynamically rebalances rare classes and adds dual reconstruction improves tail-class AP on COCO and LVIS by about 0.5 to 1.5 points in most configurations.

  4. Self-Supervised Learning for Pre-training Capsule Networks: Overcoming Medical Imaging Dataset Challenges

    cs.CV 2025-02 conditional novelty 4.0 of 10

    Contrastive learning with in-painting as a self-supervised pre-training task slightly improves capsule network accuracy on PICCOLO polyp classification (0.40 vs 0.38) but gives worse AUROC than ImageNet pre-training.

  5. Contrastive Masked Autoencoders for Character-Level Open-Set Writer Identification

    cs.CV 2025-01 conditional novelty 4.0 of 10

    A contrastive masked autoencoder achieves 89.7% precision on open-set character-level writer identification on CASIA-OLHWDB, and 81.6% rank-1 on IAM-OnDB.

  6. Object Tracking in a $360^o$ View: A Novel Perspective on Bridging the Gap to Biomedical Advancements

    cs.CV 2024-12 unverdicted novelty 2.0 of 10

    A review of object tracking algorithms for biomedical video concludes that deep learning is the most capable family, but it includes a placeholder citation for a model described as real.

Pith tools