1

Here we present a flexible algorithm, Graph Smoothing Splines Neural Networks (GSSNN), for graph classification. By employing the smoothing splines to enhance the important nodes features, our algorithm enables a high-quality and more robust graph representation. The important nodes features are served as knots that can be used for interpreting classification results.

Here we develop we develop a novel Geometry Interaction Learning (GIL) method for graphs, a well-suited and efficient alternative for learning complex geometric properties real-world graphs. Our method endows each node the freedom to determine the importance of each geometry space via a flexible dual feature interaction learning and probability assembling mechanism.