On First-Order Meta-Learning Algorithms

Benchmark Model Rank Results
few-shot-image-classification-on-mini-12Reptile+BN#11Accuracy: 32.0
few-shot-image-classification-on-mini-12Reptile#14Accuracy: 31.1
few-shot-image-classification-on-mini-13Reptile+BN#12Accuracy: 47.6
few-shot-image-classification-on-mini-13Reptile#14Accuracy: 44.7
few-shot-image-classification-on-mini-2Reptile + Transduction#91Accuracy: 49.97
few-shot-image-classification-on-mini-3Reptile + Transduction#83Accuracy: 65.99
few-shot-image-classification-on-omniglot-1-1Reptile + Transduction#17Accuracy: 89.43%
few-shot-image-classification-on-omniglot-1-2Reptile + Transduction#16Accuracy: 97.68
few-shot-image-classification-on-omniglot-5-1Reptile + Transduction#17Accuracy: 97.12%
few-shot-image-classification-on-omniglot-5-2Reptile + Transduction#13Accuracy: 99.48
few-shot-image-classification-on-tiered-2Reptile+BN#10Accuracy: 35.3
few-shot-image-classification-on-tiered-2Reptile#13Accuracy: 33.7
few-shot-image-classification-on-tiered-3Reptile+BN#12Accuracy: 52.0
few-shot-image-classification-on-tiered-3Reptile#13Accuracy: 48.0