AI Paper Atlas

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

Semi-Supervised Active Learning for Semi-Supervised Models: Exploit Adversarial Examples With Graph-Based Virtual Labels

Jiannan Guo, Haochen Shi, Yangyang Kang, Kun Kuang, Siliang Tang, Zhuoren Jiang, Changlong Sun, Fei Wu, Yueting Zhuang

ICCV 2021 · Label-efficient and robust learning › Semi-supervised learning · approach: Confidence-based pseudo-label selection · 45 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 label propagator (ablation)REVIVAL · CIFAR-100 (semi-supervised, 25% labeled)accuracy (%): 70.73 -> 71.63Improved
add virtual adversarial generator (ablation)REVIVAL · CIFAR-100 (semi-supervised, 25% labeled)accuracy (%): 70.85 -> 71.63Improved
add boundary limitator (ablation)REVIVAL · CIFAR-100 (semi-supervised, 25% labeled)accuracy (%): 71.28 -> 71.63Improved
add boundary limitator (ablation)REVIVAL · CIFAR-100 (semi-supervised, 15% labeled)accuracy (%): 67.71 -> 67.91Improved
use semi-supervised task model instead of supervised task modelREVIVAL · CIFAR-100 (25% labeled)accuracy (%): 9.33% more accuracyImproved
use REVIVAL selection instead of ICALMixMatch · CIFAR-100 (semi-supervised, 25% labeled)accuracy (%): 70.98 -> 71.63Improved
select by entropy only instead of random samplingEntropy baseline · CIFAR-10 (semi-supervised)accuracy (%)Hurt
increase kNN graph size k from 1 to 20 then beyond (ablation)REVIVAL · CIFAR-10 (supervised, 30% labeled)mean accuracy (%)Mixed
increase screening size M beyond 14500 (ablation)REVIVAL · CIFAR-10 (supervised, 30% labeled)mean accuracy (%)Mixed

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