Learning Long Range Dependencies on Graphs via Random Walks

Benchmark Model Rank Results
graph-classification-on-cifar10-100kNeuralWalker#1Accuracy (%): 80.027 ± 0.185
graph-classification-on-mnistNeuralWalker#2Accuracy: 98.760 ± 0.079
graph-classification-on-peptides-funcNeuralWalker#11AP: 0.7096 ± 0.0078
graph-regression-on-peptides-structNeuralWalker#12MAE: 0.2463 ± 0.0005
graph-regression-on-zincNeuralWalker#7MAE: 0.065 ± 0.001
link-prediction-on-pcqm-contactNeuralWalker#17MRR-ext-filtered: 0.4707 ± 0.0007
node-classification-on-clusterNeuralWalker#6Accuracy: 78.189 ± 0.188
node-classification-on-coco-spNeuralWalker#1macro F1: 0.4398 ± 0.0033
node-classification-on-pascalvoc-sp-1NeuralWalker#1macro F1: 0.4912 ± 0.0042
node-classification-on-patternNeuralWalker#4Accuracy: 86.977 ± 0.012
node-classification-on-pokecNeuralWalker#1Accuracy: 86.46 ± 0.09