AI Paper Atlas

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

Masked Siamese Networks for Label-Efficient Learning

Mahmoud Assran, Mathilde Caron, Ishan Misra, Piotr Bojanowski, Florian Bordes, Pascal Vincent, Armand Joulin, Michael Rabbat, Nicolas Ballas

ECCV 2022 · Label-efficient and robust learning › Self-supervised and contrastive · approach: Masked autoencoder pretraining · 306 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 a random mask to the anchor view (ablation)MSN · ImageNet-1K 1% labels (ViT-B/16)top-1 accuracy: 49.3 -> 52.3Improved
use only focal mask instead of no mask (ablation)MSN · ImageNet-1K 1% labels (ViT-B/16)top-1 accuracy: 49.3 -> 39.3Hurt
add focal mask views to random mask views (ablation)MSN · ImageNet-1K 1% labels (ViT-B/16)top-1 accuracy: 52.3 -> 59.8Improved
add ColorJitter to the anchor view relative to sharing the target view (ablation)MSN · ImageNet-1K 1% labels (ViT-B/16)top-1 accuracy: 7.0 -> 48.7Improved
add independent crop, flip and blur to the anchor view (ablation)MSN · ImageNet-1K 1% labels (ViT-B/16)top-1 accuracy: 48.7 -> 52.3Improved
increase random masking ratio from 0.15 to 0.5 for ViT-S/16 (ablation)MSN · ImageNet-1K 1% labels (ViT-S/16)top-1 accuracy: 66.3 -> 64.8Hurt
increase random masking ratio from 0.15 to 0.3 for ViT-B/16 (ablation)MSN · ImageNet-1K 1% labels (ViT-B/16)top-1 accuracy: 68.8 -> 69.6Improved
increase random masking ratio from 0.5 to 0.7 for ViT-L/16 (ablation)MSN · ImageNet-1K 1% labels (ViT-L/16)top-1 accuracy: 70.1 -> 69.4Hurt
increase masking ratio from 0.0 to 0.7 on ViT-L/7 (ablation)MSN · pre-training efficiencythroughput (imgs/s): 415 -> 600Improved
replace Sinkhorn normalization with tuned me-max weight 5.0 (ablation)MSN ViT-S/16 · ImageNet-1K 1% labels (ViT-S/16)top-1 accuracy: 66.4 -> 67.2Improved
remove Sinkhorn normalization without retuning me-max weight (ablation)MSN ViT-S/16 · ImageNet-1K 1% labels (ViT-S/16)top-1 accuracy: 66.4 -> 60.8Hurt
halve number of prototypes from 1024 to 512 (ablation)MSN ViT-B/16 · ImageNet-1K 1% labelstop-1 accuracy: 69.5 -> 67.6Hurt

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