HyperAI超神经
首页资讯最新论文教程数据集百科SOTALLM 模型天梯GPU 天梯顶会
全站搜索
关于
HyperAI超神经

Semi Supervised Image Classification On Cifar 2

评估指标

Percentage error

评测结果

各个模型在此基准测试上的表现结果

模型名称
Percentage error
Paper TitleRepository
SMPL (WRN-28-8)21.68Self Meta Pseudo Labels: Meta Pseudo Labels Without The Teacher-
Dash (RA, WRN-28-8)21.97±0.14Dash: Semi-Supervised Learning with Dynamic Thresholding-
CCSSL(FixMatch)19.32Class-Aware Contrastive Semi-Supervised Learning
FlexMatch21.90±0.15FlexMatch: Boosting Semi-Supervised Learning with Curriculum Pseudo Labeling
FreeMatch21.68FreeMatch: Self-adaptive Thresholding for Semi-supervised Learning
Ⅱ-Model39.19Regularization With Stochastic Transformations and Perturbations for Deep Semi-Supervised Learning-
SimPLE (WRN-28-8)21.89SimPLE: Similar Pseudo Label Exploitation for Semi-Supervised Classification
SHOT-VAE25.3SHOT-VAE: Semi-supervised Deep Generative Models With Label-aware ELBO Approximations
R2-D2 (CNN-13)32.87Repetitive Reprediction Deep Decipher for Semi-Supervised Learning
LaplaceNet (WRN-28-8)22.11± 0.23LaplaceNet: A Hybrid Graph-Energy Neural Network for Deep Semi-Supervised Classification
SimMatch20.58SimMatch: Semi-supervised Learning with Similarity Matching
Dual Student (480)32.77Dual Student: Breaking the Limits of the Teacher in Semi-supervised Learning
DP-SSL22.24±0.31DP-SSL: Towards Robust Semi-supervised Learning with A Few Labeled Samples-
SemCo (μ=7)24.45±0.12All Labels Are Not Created Equal: Enhancing Semi-supervision via Label Grouping and Co-training
FixMatch (CTA, WRN-28-8)23.18±0.11FixMatch: Simplifying Semi-Supervised Learning with Consistency and Confidence
Temporal ensembling38.65Temporal Ensembling for Semi-Supervised Learning
NP-Match21.22NP-Match: When Neural Processes meet Semi-Supervised Learning
CowMix (WRN-28-96x2d)23.07±0.30Milking CowMask for Semi-Supervised Image Classification
EnAET (WRN-28-2)26.93±0.21EnAET: A Self-Trained framework for Semi-Supervised and Supervised Learning with Ensemble Transformations
FixMatch+CR21.03Contrastive Regularization for Semi-Supervised Learning-
0 of 27 row(s) selected.