Shot in the Dark: Few-Shot Learning with No Base-Class Labels

Benchmark Model Rank Results
unsupervised-few-shot-image-classification-on-mini-imagenet-5-way-1-shotUBC-FSL–Accuracy: 57.1
unsupervised-few-shot-image-classification-on-mini-imagenet-5-way-5-shotUBC-FSL–Accuracy: 77.2
unsupervised-few-shot-image-classification-on-tiered-imagenet-5-way-1-shotUBC-FSL–Accuracy: 68.0
unsupervised-few-shot-image-classification-on-tiered-imagenet-5-way-5-shotUBC-FSL–Accuracy: 84.3