AI Paper Atlas

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

Tangent-Normal Adversarial Regularization for Semi-Supervised Learning

Bing Yu, Jingfeng Wu, Jinwen Ma, Zhanxing Zhu

CVPR 2019 · Label-efficient and robust learning › Semi-supervised learning · approach: GAN-based semi-supervised learning · 29 citations (OpenAlex)

Check 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
Use tangent and normal adversarial regularization together (ablation)TNAR-VAE · CIFAR-10, 4,000 labels, no augmentationclassification error (%): 13.87\pm 0.32 -> 12.06\pm 0.35; 15.91\pm 0.09 -> 12.06\pm 0.35Improved
Add tangent adversarial regularization to normal-only regularization (ablation)TNAR-VAE · FashionMNIST, 100 labelsclassification error (%): 27.83 -> 23.35Improved
Add normal adversarial regularization to tangent-only regularization (ablation)TNAR-VAE · FashionMNIST, 100 labelsclassification error (%): 26.45 -> 23.35Improved
Replace the VAE-estimated manifold with a localized-GAN-estimated manifoldTNAR · FashionMNIST, 100 labelsclassification error (%): 23.35 -> 23.65Hurt
Use TNAR with an oracle manifold rather than VATTwo-rings artificial dataset · Two-rings artificial datasetclassification error (%): 23.80 -> 0Improved
Use TNAR-VAE rather than VATFashionMNIST · FashionMNIST, 1,000 labelsclassification error (%): 14.51 -> 12.86Improved
Use TNAR-VAE rather than VAT (large)TNAR-VAE (large) · SVHN, 1,000 labels, no augmentationclassification error (%): 4.28\pm 0.10 -> 3.80\pm 0.12Improved
Use TNAR-VAE rather than VAT (large)TNAR-VAE (large) · CIFAR-10, 4,000 labels, with augmentationclassification error (%): 10.55\pm 0.05 -> 8.85\pm 0.03Improved
Add entropy regularization (ablation)TNAR-Manifold · Two-rings artificial datasetclassification error (%): 9.90 -> 0Improved
Add entropy regularization (ablation)VAT · Two-rings artificial datasetclassification error (%)Hurt

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