To understand what an autoencoder has learned through its unsupervised learning process. To rank hidden nodes according to their capability of performing a learning task. To identify the specialty ...

The basic idea of the proposed method is to eliminate lazy nodes which rarely affect the model performance based on the weight distribution of the bottleneck layer. Since the proposed method takes ...

Abstract: The purpose of graph embedding is to encode the known node features ... with a large iterative learning time, and susceptible to local optimal solutions. Thus, we propose Graph Convolutional ...

This is a Keras implementation of the symmetrical autoencoder architecture with parameter sharing for the tasks of link prediction and semi-supervised node classification, as described in the ...