A Unified Deep Learning Formalism For Processing Graph Signals
This work addresses the challenge of graph signal processing for researchers in machine learning, but it is incremental as it reviews and unifies existing methods rather than introducing new ones.
The paper tackles the problem of extending convolutional neural networks to process signals on arbitrary graphs by reviewing and unifying existing deep learning models into a single formalism, providing a comparative analysis.
Convolutional Neural Networks are very efficient at processing signals defined on a discrete Euclidean space (such as images). However, as they can not be used on signals defined on an arbitrary graph, other models have emerged, aiming to extend its properties. We propose to review some of the major deep learning models designed to exploit the underlying graph structure of signals. We express them in a unified formalism, giving them a new and comparative reading.