Extended Few-Shot Learning: Exploiting Existing Resources for Novel Tasks

Benchmark Model Rank Results
few-shot-image-classification-on-fc100-5-way-1-shotpseudo-shots#4Accuracy: 50.57
few-shot-image-classification-on-cifar-fs-5-way-1-shotpseudo-shots#11Accuracy: 81.87
few-shot-image-classification-on-cifar-fs-5-way-5-shotpseudo-shots#13Accuracy: 89.12
few-shot-image-classification-on-fc100-5-way-5-shotpseudo-shots#13Accuracy: 61.58
few-shot-image-classification-on-tiered-imagenet-5-way-1-shotpseudo-shots#16Accuracy: 76.55
few-shot-image-classification-on-tiered-imagenet-5-way-5-shotpseudo-shots#21Accuracy: 86.82
few-shot-image-classification-on-mini-imagenet-5-way-1-shotpseudo-shots#26Accuracy: 73.35
few-shot-image-classification-on-mini-imagenet-5-way-5-shotpseudo-shots#39Accuracy: 82.51