I'm new to the area, and we don't have a course on graph neural networks at our university. However, I will still like to know the main theoretical results when considering convergence of graph neural networks. Where can I learn more about the topic? Thanks.


1 Answer 1


More generally, this area is known as geometric deep learning, i.e., it encompasses learning not only on graph but on other non-Euclidean domains. A good broad start is the survey of Bronstein et al. [1], after which I can recommend basically any material by Bronstein like his many excellent presentations (also available on Youtube - just do a search).

For graph neural nets, you might start with Kipf & Welling (the survey [1] will give references). Perhaps loosely building on their work, there is also GraphSAGE [2] and later developments like graph attention networks. I've not actively followed on the literature for the past 2-3 years, so I don't know what's hot currently or how badly outdated these methods might be, but these works definitely will give you the foundation.

[1] Bronstein, Michael M., Joan Bruna, Yann LeCun, Arthur Szlam, and Pierre Vandergheynst. "Geometric deep learning: going beyond Euclidean data." IEEE Signal Processing Magazine 34, no. 4 (2017): 18-42.

[2] Hamilton, William L., Rex Ying, and Jure Leskovec. "Inductive representation learning on large graphs." In Proceedings of the 31st International Conference on Neural Information Processing Systems, pp. 1025-1035. 2017.

  • $\begingroup$ Thanks! Are there materials that are structured like a course, or a series of videos? $\endgroup$
    – nir shahar
    Jul 5, 2021 at 10:45
  • $\begingroup$ @nirshahar The ones I'm thinking of are more like conference (keynote) talks, so high level intros to main results etc. $\endgroup$
    – Juho
    Jul 5, 2021 at 12:25

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.