Semi Supervised Image Classification On Svhn

المقاييس

Accuracy

النتائج

نتائج أداء النماذج المختلفة على هذا المعيار القياسي

		Paper Title
Meta Pseudo Labels (WRN-28-2)	98.01 ± 0.07	Meta Pseudo Labels
DoubleMatch	97.90 ± 0.07	DoubleMatch: Improving Semi-Supervised Learning with Self-Supervision
FixMatch (CTA)	97.64±0.19	FixMatch: Simplifying Semi-Supervised Learning with Consistency and Confidence
EnAET	97.58	EnAET: A Self-Trained framework for Semi-Supervised and Supervised Learning with Ensemble Transformations
UDA	97.54	Unsupervised Data Augmentation for Consistency Training
ReMixMatch	97.17	ReMixMatch: Semi-Supervised Learning with Distribution Alignment and Augmentation Anchoring
MixMatch	96.73	MixMatch: A Holistic Approach to Semi-Supervised Learning
Triple-GAN-V2 (CNN-13)	96.55	Triple Generative Adversarial Networks
ICT (WRN-28-2)	96.47	Interpolation Consistency Training for Semi-Supervised Learning
R2-D2 (CNN-13)	96.36	Repetitive Reprediction Deep Decipher for Semi-Supervised Learning
FCE	96.13	Flow Contrastive Estimation of Energy-Based Models
ICT	96.11	Interpolation Consistency Training for Semi-Supervised Learning
Mean Teacher	96.05	Mean teachers are better role models: Weight-averaged consistency targets improve semi-supervised deep learning results
Triple-GAN-V2 (CNN-13, no aug)	96.04	Triple Generative Adversarial Networks
VAT	94.58	Virtual Adversarial Training: A Regularization Method for Supervised and Semi-Supervised Learning
SESEMI SSL (ConvNet)	94.41	Exploring Self-Supervised Regularization for Supervised and Semi-Supervised Learning
GAN	91.89	Improved Techniques for Training GANs

0 of 17 row(s) selected.

Semi Supervised Image Classification On Svhn | SOTA | HyperAI