Natural Graph Networks
| Authors | |
|---|---|
| Publication date | 2021 |
| Host editors |
|
| Book title | 34th Concerence on Neural Information Processing Systems (NeurIPS 2020) |
| Book subtitle | online, 6-12 December 2020 |
| ISBN |
|
| Series | Advances in Neural Information Processing Systems |
| Event | Advances in Neural Information Processing Systems 2020 |
| Volume | Issue number | 5 |
| Pages (from-to) | 3636-3646 |
| Publisher | San Diego, CA: Neural Information Processing Systems Foundation |
| Organisations |
|
| Abstract |
A key requirement for graph neural networks is that they must process a graph in
a way that does not depend on how the graph is described. Traditionally this has been taken to mean that a graph network must be equivariant to node permutations. Here we show that instead of equivariance, the more general concept of naturality is sufficient for a graph network to be well-defined, opening up a larger class of graph networks. We define global and local natural graph networks, the latter of which are as scalable as conventional message passing graph neural networks while being more flexible. We give one practical instantiation of a natural network on graphs which uses an equivariant message network parameterization, yielding good performance on several benchmarks. |
| Document type | Conference contribution |
| Note | With supplemental file |
| Language | English |
| Published at | https://papers.nips.cc/paper/2020/hash/2517756c5a9be6ac007fe9bb7fb92611-Abstract.html |
| Other links | https://www.proceedings.com/59066.html |
| Downloads |
NeurIPS-2020-natural-graph-networks-Paper
(Accepted author manuscript)
|
| Supplementary materials | |
| Permalink to this page | |