by
Alejandro Ribeiro and Fernando Gama
University of Pennsylvania

Due to their remarkable success, convolutional neural networks (CNNs) have become the de-facto standard for machine learning in time and images. Over the last few years we have seen the emergence of graph neural networks (GNNs) as a generalization of CNNs for the processing of information that is supported on graphs. GNNs are rapidly becoming the tool of choice for machine learning and signal processing on graphs and networks. In this tutorial we will cover architectures, fundamental properties, and applications of GNNs. This results in a tutorial that is divided in the following three modules:

  1. Graph Neural Network Architectures: Convolutions on Graphs.
  2. Fundamental Properties of Graph Neural Networks: Stability and Discriminability.
  3. Applications of Graph Neural Networks: Multi-Agent Systems.
%d bloggers like this: