AI Paper Atlas

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

Preservational Learning Improves Self-Supervised Medical Image Models by Reconstructing Diverse Contexts

Hong-Yu Zhou, Chixiang Lu, Sibei Yang, Xiaoguang Han, Yizhou Yu

ICCV 2021 · Label-efficient and robust learning › Self-supervised and contrastive · approach: Inductive bias and minimal representations · 114 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

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 self-reconstruction branch (ablation)ContraLoss · Chest14->Chest14 (9:1)AUC: 71.7 -> 72.5Improved
add TransAtt (with rotation) to contrastive loss (ablation)ContraLoss · Chest14->Chest14 (9:1)AUC: 71.7 -> 74.4Improved
add CrossMix to contrastive loss (ablation)ContraLoss · Chest14->Chest14 (9:1)AUC: 71.7 -> 73.7Improved
add rotation transformations to TransAtt indicator (beyond flip) (ablation)TransAtt · Chest14->Chest14 (9:1)AUC: 73.5 -> 74.4Improved
replace self-reconstruction with rotation prediction (ablation)ContraLoss + pretext · Chest14->Chest14 (9:1)AUC: 72.5 -> 72.3Hurt
halve CrossMix alpha from 1 to 0.5 (ablation)CrossMix · Chest14->Chest14 (9:1)AUC: 73.7 -> 73.3Hurt
combine TransAtt and CrossMix into full PCRL (ablation)ContraLoss · Chest14->Chest14 (8:2)AUC: 74.8 -> 78.8Improved
add TransAtt over self-reconstruction on rotation pretext task (ablation)ContraLoss + Self-Recons. · Chest14 rotation predictionaccuracy: 87.9 -> 91.0Improved
add TransAtt over self-reconstruction on position pretext task (ablation)ContraLoss + Self-Recons. · Chest14 relative position predictionaccuracy: 84.5 -> 89.3Improved
add CrossMix to contrastive loss on position pretext (ablation)ContraLoss · Chest14 relative position predictionaccuracy: 82.3 -> 88.3Improved
use PCRL instead of MoCov2 for ImageNet pretrainingself-supervised pretraining · CityscapesmIoU: 76.3 -> 77.3Improved
use PCRL instead of MoCov2 for ImageNet pretrainingself-supervised pretraining · COCOmAP: 40.5 -> 41.3Improved

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