List your company
ResearchFormerly WizWand

5-way 5-shot classification on miniImageNet (test)

86.61AccuracyS2M2 + TCPR
54.047662.501370.95579.4087Aug 19, 2021Oct 30, 2021Jan 11, 2022Mar 25, 2022Jun 6, 2022Aug 18, 2022Oct 30, 2022
Updated 8mo ago

Evaluation results

MethodLinks
S2M2 + TCPRBackbone=WRN-28, 9-patch cropping=true2022.1086.61
S2M2Backbone=WRN-28, 9-patch cropping=true2022.1085.63
Inv-Equ + TCPRBackbone=ResNet-122022.1085.32
COSOCBackbone=ResNet-122022.1085.16
S2M2 + TCPRBackbone=WRN-282022.1084.51
Inv-EquBackbone=ResNet-122022.1084.43
PALBackbone=ResNet-122022.1084.4
PALBackbone=ResNet-122022.1084.4
DeepEMDBackbone=ResNet-12, 9-patch cropping=true2022.1084.13
POODLEBackbone=ResNet-122022.1083.5
S2M2Backbone=WRN-282022.1083.5
MELRBackbone=ResNet-122022.1083.4
IEPTBackbone=ResNet-122022.1082.9
DCBackbone=WRN-282022.1082.88
DMFBackbone=ResNet-122022.1082.71
S/TBackbone=ResNet-122022.1082.53
InfoPatchBackbone=ResNet-122022.1082.44
DeepEMDBackbone=ResNet-122022.1082.41
DeepEMDBackbone=ResNet-122022.1082.41
BaseTransformerBackbone=ResNet-122022.1082.37
FEATBackbone=ResNet-122022.1082.05
CSEIBackbone=ResNet-122022.1081.93
METALBackbone=ResNet-122022.1081.43
META-QDABackbone=ResNet-122022.1080.98
Baseline++ + TCPRBackbone=ResNet-122022.1080.87
ArLBackbone=ResNet-122022.1080.41
SimpleShotBackbone=ResNet-122022.1080.02
CANBackbone=ResNet-122022.1079.44
Baseline++Backbone=ResNet-122022.1079.32
ProtoNetsBackbone=ResNet-122022.1078.02
EGNN + TMDlayer + TransductionTransduction=Yes2021.0877.78
Baseline++ + TCPRBackbone=ResNet-182022.1077.31
EGNN + TransductionTransduction=Yes2021.0876.37
Baseline++Backbone=ResNet-182022.1076.11
IEPTBackbone=Conv4-642022.1073.91
ProtoNetBackbone=ResNet-182022.1073.68
BaseTransformerBackbone=Conv4-642022.1073.4
META-QDABackbone=Conv4-642022.1072.64
MELRBackbone=Conv4-642022.1072.27
FEATBackbone=Conv4-642022.1071.61
TPN (Higher K)Transduction=Yes2021.0869.86
TNPTransduction=Yes2021.0869.43
MatchingNetBackbone=ResNet-182022.1068.88
EGNN + TMDlayerTransduction=No2021.0868.35
ProtoNetsBackbone=Conv4-642022.1068.2
Relation NetTransduction=BN2021.0867.07
SimpleShotBackbone=Conv4-642022.1066.92
EGNNTransduction=No2021.0866.85
GNNTransduction=No2021.0866.41
MAML + TransductionTransduction=Yes2021.0866.19
Reptile + BNTransduction=BN2021.0865.99
Prototypical NetTransduction=No2021.0865.77
MAMLBackbone=ResNet-182022.1065.72
MAMLTransduction=BN2021.0863.11
ReptileTransduction=No2021.0862.74
Matching NetworksTransduction=No2021.0855.3