PDO-eConvs: Partial Differential Operator Based Equivariant Convolutions

Open paper
Benchmark Model Rank Results
image-classification-on-cifar-10PDO-eConv (p8, 4.6M)#133Percentage correct: 96.5
image-classification-on-cifar-10PDO-eConv (p8, 2.62M)#140Percentage correct: 96.32
image-classification-on-cifar-10PDO-eConv (p6m,0.37M)#178Percentage correct: 94.62
image-classification-on-cifar-10PDO-eConv (p6,0.36M)#186Percentage correct: 94.35
image-classification-on-cifar-100PDO-eConv (p8, 4.6M)#144Percentage correct: 81.6
image-classification-on-cifar-100PDO-eConv (p8, 2.62M)#163Percentage correct: 79.99
image-classification-on-cifar-100PDO-eConv (p6m,0.37M)#199Percentage correct: 73
image-classification-on-cifar-100PDO-eConv (p6,0.36M)#202Percentage correct: 72.87

Rank counts only results with a code link.