Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

fMNIST

Benchmarks

Task NameDataset NameSOTA ResultTrend
Image ClassificationFMNIST (test)
Test Accuracy90.9
244
ClassificationFMNIST (test)
Accuracy99.06
152
Backdoor AttackFMNIST
ASR100
75
Image ClassificationFMNIST 0.5 (test)
Accuracy78.06
72
ClassificationFMNIST
Accuracy84.28
53
Backdoor Attack in Dataset CondensationFMNIST
CTA88.4
43
Dimensionality ReductionFMNIST
AUC R_NX Score76.48
42
Image ClassificationFMNIST
Accuracy89.23
39
Image ClassificationFMNIST
Accuracy (IID)84.842
33
ClusteringFMNIST
NMI66.35
31
Anomaly DetectionFMNIST
Avg AUROC0.956
29
Image ClassificationFMNIST ⊥
GM Score87.54
27
Out-of-Distribution DetectionFMNIST
OOD Score97.3
26
Image ClassificationFMNIST
Speedup69.24
21
Dimensionality ReductionFMNIST
AUC RNX0.7648
21
Stealthiness AssessmentFMNIST
PSNR60.11
20
Image ClassificationFMNIST
Accuracy92.54
20
KNN ClassificationFMNIST
Accuracy (k=1)87.47
20
Silhouette Score EvaluationFMNIST
Silhouette Score0.2315
20
KNN ClassificationFMNIST
Accuracy (k=5)88.3
20
Image ClassificationRotated FMNIST
Accuracy89.9
18
OOD DetectionFMNIST In-distribution: CIFAR100 (test)
AUPR (%)78.6
18
OOD DetectionFMNIST In-distribution: CIFAR10 (test)
AUPR92.5
18
Untargeted white-box adversarial attackFMNIST original (test)
Original Error Rate0.1
16
Image ClassificationFMNIST (Shards)
RES100
14
Showing 25 of 82 rows