AI Paper Atlas

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

Compressive Visual Representations

Kuang-Huei Lee, Anurag Arnab, Sergio Guadarrama, John Canny, Ian Fischer

NeurIPS 2021 · Label-efficient and robust learning › Self-supervised and contrastive · approach: Empirical contrastive visual learning · 2 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 CEB compression to SimCLRSimCLR · ImageNet, ResNet-50, 1000 epochsTop-1 accuracy: 70.7 -> 71.6Improved
Add CEB compression to BYOLBYOL · ImageNet, ResNet-50, 1000 epochsTop-1 accuracy: 74.2 -> 75.6Improved
Add CEB compression to BYOLBYOL · ImageNet-CTop-1 accuracy: 42.7 -> 45.1Improved
Add CEB compression to SimCLRSimCLR · ImageNet, 1% labelsTop-1 accuracy: 49.3 -> 51.1Improved
Add CEB compression to BYOLBYOL · ImageNet, 1% labelsTop-1 accuracy: 55.5 -> 60.6Improved
Increase compression strength to β = 1 (ablation)C-SimCLR · ImageNet, 300 epochsTop-1 accuracy: 69.0 -> 70.2Improved
Use β = 0 (no explicit compression) with a stochastic representation (ablation)C-SimCLR · ImageNet, 300 epochsTop-1 accuracy: 69.0 -> 69.7Improved
Increase compression strength beyond β = 1 to β = 1.5 (ablation)C-SimCLR · ImageNet, 300 epochsTop-1 accuracy: 70.2 -> 69.7Hurt
Increase compression strength to β = 2 (ablation)C-SimCLR · ImageNettraining stabilityHurt
Increase BYOL loss weight from 1.0 to 5.0 (ablation)BYOL · ImageNet, 300 epochsTop-1 accuracy: 72.5 -> 72.8Improved
Increase BYOL loss weight from 1.0 to 5.0 (ablation)BYOL · ImageNet, 1000 epochsTop-1 accuracy: 74.2 -> 74.2No clear effect
Add a 256-d linear layer and l2-normalization (ablation)BYOL · ImageNet, 300 epochsTop-1 accuracy: 72.8 -> 72.8No 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