On the unreasonable effectiveness of feature propagation in learning on graphs with missing node features
While Graph Neural Networks (GNNs) have recently become the de facto standard for modeling relational data, they impose a strong assumption on the availability of the node or edge features of the graph. In many real-world applications, however, features are only partially available; for example, in...
Main Authors: | Rossi, E, Kenlay, H, Gorinova, MI, Chamberlain, BP, Dong, X, Bronstein, M |
---|---|
Format: | Conference item |
Sprog: | English |
Udgivet: |
Proceedings of Machine Learning Research
2022
|
Lignende værker
-
Listening to Unreason: Foucault and Wittgenstein on Reason and the Unreasonable Man
af: Liat Lavi
Udgivet: (2018-10-01) -
The age of unreason /
af: 258599 Handy, Charles B.
Udgivet: (1989) -
The age of unreason /
af: 258599 Handy, Charles B.
Udgivet: (1991) -
Robustness analysis of graph-based machine learning
af: Kenlay, H
Udgivet: (2022) -
CTCs 2020: Great Expectations or Unreasonable Dreams
af: Elisabetta Rossi, et al.
Udgivet: (2019-08-01)