Fig. 6From: Learning compact graph representations via an encoder-decoder networkThe performance of our proposed method under various settings. a Performance of various aggregation methods, b Accuracy versus training epochs, c Accuracy versus number of samples used for aggregationBack to article page