AI Paper Atlas

Research map / Label-efficient and robust learning / Semi-supervised learning

SemPPL: Predicting Pseudo-Labels for Better Contrastive Representations

Matko Bošnjak, Pierre Harvey Richemond, Nenad Tomasev, Florian Strub, Jacob C Walker, Felix Hill, Lars Holger Buesing, Razvan Pascanu, Charles Blundell, Jovana Mitrovic

ICLR 2023 · Label-efficient and robust learning › Semi-supervised learning · approach: Confidence-based pseudo-label selection · 4 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 pseudo-label-based semantic positives to ReLICv2ReLICv2 on ImageNet with 1% labels · ImageNet test, 1% labelstop-1 accuracy: 58.1 -> 68.5; +10.4%Improved
Add pseudo-label-based semantic positives to BYOLBYOL on ImageNet with 1% and 10% labels · ImageNet test, 1% and 10% labelstop-1 accuracy: +3.9%; +3.6%Improved
Use pseudo-labels to select semantic positives for unlabelled data (ablation)SemPPL on ImageNet · ImageNet test, 1% and 10% labelstop-1 accuracy: 61.9 -> 68.5; 71.1 -> 76.0Improved
Add semantic positives for labelled data (ablation)SemPPL on ImageNet · ImageNet test, 1% and 10% labelstop-1 accuracy: 58.1 -> 68.5; 72.4 -> 76.0Improved
Replace predicted pseudo-labels with ground-truth labels for positive retrieval (ablation)SemPPL on ImageNet · ImageNet test, 10% labels, 100 epochstop-1 accuracy: 69.9 -> 71.6; 1.7%Improved
Increase semantic positives from 1 or 2 to 3 (ablation)SemPPL on ImageNet · ImageNet test, 10% labelstop-1 accuracy: 69.9 -> 69.9No clear effect
Increase semantic positives to 5 (ablation)SemPPL on ImageNet · ImageNet test, 10% labelstop-1 accuracy: 69.9 -> 71.0Improved
Train for 500 rather than 300 epochs (ablation)SemPPL on ImageNet · ImageNet test, 10% labelstop-1 accuracy: 72.7 -> 72.3Hurt
Use majority voting across views for pseudo-labels (ablation)SemPPL on ImageNet · ImageNet test, 10% labelstop-1 accuracy: 69.0 -> 69.9Improved
Increase k-NN from 1 to larger neighbour counts (ablation)SemPPL on ImageNet · ImageNet test, 10% labelstop-1 accuracy: 70.0 -> 69.9 (k=2); 69.8 (k=3); 69.8 (k=5); 69.1 (k=10)Hurt
Increase labelled-embedding queue capacity (ablation)SemPPL on ImageNet · ImageNet test, 10% labelstop-1 accuracy: 67.9 -> 69.9Improved
Use SemPPL instead of the strongest listed semi-supervised baselineResNet-50 1× on ImageNet · ImageNet test, 1% labelstop-1 accuracy: 67.2 -> 68.5; +1.3%Improved

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