Probabilistically Rewired Message-Passing Neural Networks

arXiv:2310.02156v4 Announce Type: replace
Abstract: Message-passing graph neural networks (MPNNs) emerged as powerful tools for processing graph-structured input. However, they operate on a fixed input graph structure, ignoring potential noise and missing information. Furthermore, their local aggregation mechanism can lead to problems such as over-squashing and limited expressive power in capturing relevant graph structures. Existing solutions to these challenges have primarily relied on heuristic methods, often disregarding the underlying data distribution. Hence, devising principled approaches for learning to infer graph …

abstract aggregation arxiv cs.lg cs.ne graph graph neural networks however information networks neural networks noise power processing tools type