Learning to Remember Rare Events

Benchmark Model Rank Results
few-shot-image-classification-on-omniglot-1-1ConvNet with Memory Module#13Accuracy: 95%
few-shot-image-classification-on-omniglot-1-2ConvNet with Memory Module#12Accuracy: 98.4
few-shot-image-classification-on-omniglot-5-1ConvNet with Memory Module#11Accuracy: 98.6%
few-shot-image-classification-on-omniglot-5-2ConvNet with Memory Module#10Accuracy: 99.6