AI Paper Atlas

Research map / Label-efficient and robust learning / Self-supervised and contrastive

OPERA: Omni-Supervised Representation Learning with Hierarchical Supervisions

Chengkun Wang, Wenzhao Zheng, Zheng Zhu, Jie Zhou, Jiwen Lu

ICCV 2023 · Label-efficient and robust learning › Self-supervised and contrastive · approach: Invariance-based self-supervised learning · 7 citations (OpenAlex)

This paper tried a common change: Train longer

increase pretraining epochs · ImageNet finetune and ADE20K · accuracy and mIoU: 78.7 at 50 PT epochs (finetune) Improved

What 387 accepted papers report for this change

See this change in detailMatched by the wording of each paper's own result, read by AI from the full text. Papers report gains more often than failures, so this shows what gets published, not the chance a change works.

Read the paper

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add OPERA hierarchical supervision to MoCo-v3 (150 epochs vs MoCo-v3 300 epochs)MoCo-v3 ResNet-50 · ImageNet linear probetop-1 accuracy: 70.5 -> 73.7Improved
add OPERA to MoCo-v3MoCo-v3 ViT-B (BS 4096) · ImageNet finetunetop-1 accuracy: 83.0 -> 83.5Improved
add OPERA to MoCo-v3MoCo-v3 ViT-S · Oxford Flowers-102 transfertop-1 accuracy: 90.3 -> 95.5Improved
add OPERA to MoCo-v3 on ADE20K segmentationMoCo-v3 ViT-S · ADE20KmIoU: 42.3 -> 43.6Improved
impose full supervision directly on backbone feature (arrangement A) instead of the OPERA proxy hierarchy (ablation)OPERA · ImageNet / ADE20K ViT-Sclassification vs segmentationMixed
increase pretraining epochs (ablation)OPERA ViT-S · ImageNet finetune and ADE20Kaccuracy and mIoU: 78.7 at 50 PT epochs (finetune)Improved
increase number of MLP layers beyond 1 in the final block (ablation)OPERA ViT-S · ImageNettop-1 accuracyHurt
go from 1 to 2 MLP layers (ablation)OPERA ViT-S · ADE20KmIoU: +0.5 mIoUImproved
enlarge hidden dimension of MLP (ablation)OPERA ViT-S · ImageNet and ADE20Kaccuracy and mIoUNo clear effect
use OPERA instead of supervised pretraining with MLP projector (ablation)ViT-S supervised · ImageNet / ADE20Ktop-1 accuracy / mIoU: 78.4 -> 78.8 top-1; 41.9 -> 42.4 mIoUImproved
add MLP block to supervised pretraining (ablation)Supervised ViT-S · ImageNet / ADE20Ktop-1 accuracy / mIoU: 78.7 -> 78.4 top-1; 41.5 -> 41.9 mIoUMixed
add OPERA to MoCo-v3 at 150 epochs on COCO 1xMoCo-v3 R50 · COCO 1xAP box: 38.9 -> 38.9No clear effect

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy