Conference item
Beltrami flow and neural diffusion on graphs
- Abstract:
- We propose a novel class of graph neural networks based on the discretised Beltrami flow, a non-Euclidean diffusion PDE. In our model, node features are supplemented with positional encodings derived from the graph topology and jointly evolved by the Beltrami flow, producing simultaneously continuous feature learning and topology evolution. The resulting model generalises many popular graph neural networks and achieves state-of-the-art results on several benchmarks.
- Publication status:
- Published
- Peer review status:
- Peer reviewed
Actions
Access Document
- Files:
-
-
(Preview, Accepted manuscript, pdf, 1.8MB, Terms of use)
-
Authors
- Publisher:
- Curran Associates
- Host title:
- Advances in Neural Information Processing Systems 34
- Volume:
- 3
- Pages:
- 1594-1609
- Publication date:
- 2022-05-03
- Acceptance date:
- 2021-09-28
- Event title:
- 35th Conference on Neural Information Processing Systems (NeurIPS 2021)
- Event location:
- Virtual event
- Event website:
- https://nips.cc/Conferences/2021
- Event start date:
- 2021-12-14
- Event end date:
- 2021-12-16
- ISSN:
-
1049-5258
- ISBN:
- 9781713845393
- Language:
-
English
- Keywords:
- Pubs id:
-
1265689
- Local pid:
-
pubs:1265689
- Deposit date:
-
2022-09-13
- ARK identifier:
Terms of use
- Copyright holder:
- Chamberlain et al. and NIPS
- Copyright date:
- 2021
- Rights statement:
- Copyright © 2019 by the authors and NIPS.
- Notes:
- This is the accepted manuscript version of the paper. The final version is available online at: https://proceedings.neurips.cc/paper/2021/hash/0cbed40c0d920b94126eaf5e707be1f5-Abstract.html
If you are the owner of this record, you can report an update to it here: Report update to this record