Lesson 21 of 21 · Part V, Deep learning and beyond

Learning on structured data and graphs

From vectors to graphs: transductions on graphs, message passing and Deep Graph Networks (GCN, NN4G, GNN, GraphESN), the problems of depth (over-smoothing, over-squashing, heterophily) and kernels for structures.

12 figures to manipulate.

In this lesson

• Why structured data and graphs
• Deep Graph Networks
• Open problems and some proposals
• Other approaches: kernels for structures

Other lessons

Recurrent neural networks (RNN) · All lessons

Notes by Fabio Piscitelli — Machine Learning (654AA), Prof. Alessio Micheli, University of Pisa, a.y. 2026/27.