AI Paper Atlas

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

Learning Generalized Transformation Equivariant Representations Via AutoEncoding Transformations

Guo-Jun Qi, Liheng Zhang, Feng Lin, Xiao Wang

TPAMI 2020 · Label-efficient and robust learning › Self-supervised and contrastive · approach: Invariance-based self-supervised learning · 47 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
replace deterministic AET with probabilistic AVTAET-project · CIFAR-10 conv classifiererror rate: 7.82 -> 7.75Improved
replace affine with projective transformationsAET · CIFAR-10 conv classifiererror rate: 8.05 -> 7.82Improved
replace RotNet with AET-projectRotNet + conv · CIFAR-10error rate: 8.84 -> 7.82Improved
replace RotNet with AVT-projectRotNet · CIFAR-10 KNN K=5error rate: 25.01 -> 21.62Improved
use AVT instead of RotNet with few labels (1000 per class)RotNet + conv · CIFAR-10 few labelserror rate: 13.57 -> 12.27Improved
use AVT with 20 labels per class vs RotNetRotNet + conv · CIFAR-10 20 labels/classerror rate: 35.37 -> 35.44No clear effect
replace RotNet with AVT-projectRotNet · ImageNet Conv5 non-lineartop-1 accuracy: 43.8 -> 48.4Improved
replace AET with AVT-projectAET-project · ImageNet linear Conv5top-1 accuracy: 37.7 -> 39.1Improved
replace RotNet with AVT-projectRotNet · Places Conv5top-1 accuracy: 33.7 -> 35.6Improved
replace Mean Teacher with SATMean Teacher · CIFAR-10 4000 labelserror rate: 12.31 -> 9.58Improved
replace Mean Teacher with SATMean Teacher · SVHN 250 labelserror rate: 4.35 -> 4.30No clear effect
add entropy minimization (ablation)SAT · CIFAR-10 4000 labelserror rate: 10.90 -> 9.58Improved

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