Gumbel-MPNN: Graph Rewiring with Gumbel-Softmax
Fuente:
arXiv
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866914002510020608 |
|---|---|
| author | Hoffmann, Marcel Galke, Lukas Scherp, Ansgar |
| author_facet | Hoffmann, Marcel Galke, Lukas Scherp, Ansgar |
| contents | Graph homophily has been considered an essential property for message-passing neural networks (MPNN) in node classification. Recent findings suggest that performance is more closely tied to the consistency of neighborhood class distributions. We demonstrate that the MPNN performance depends on the number of components of the overall neighborhood distribution within a class. By breaking down the classes into their neighborhood distribution components, we increase measures of neighborhood distribution informativeness but do not observe an improvement in MPNN performance. We propose a Gumbel-Softmax-based rewiring method that reduces deviations in neighborhood distributions. Our results show that our new method enhances neighborhood informativeness, handles long-range dependencies, mitigates oversquashing, and increases the classification performance of the MPNN. The code is available at https://github.com/Bobowner/Gumbel-Softmax-MPNN. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2508_17531 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Gumbel-MPNN: Graph Rewiring with Gumbel-Softmax Hoffmann, Marcel Galke, Lukas Scherp, Ansgar Machine Learning Graph homophily has been considered an essential property for message-passing neural networks (MPNN) in node classification. Recent findings suggest that performance is more closely tied to the consistency of neighborhood class distributions. We demonstrate that the MPNN performance depends on the number of components of the overall neighborhood distribution within a class. By breaking down the classes into their neighborhood distribution components, we increase measures of neighborhood distribution informativeness but do not observe an improvement in MPNN performance. We propose a Gumbel-Softmax-based rewiring method that reduces deviations in neighborhood distributions. Our results show that our new method enhances neighborhood informativeness, handles long-range dependencies, mitigates oversquashing, and increases the classification performance of the MPNN. The code is available at https://github.com/Bobowner/Gumbel-Softmax-MPNN. |
| title | Gumbel-MPNN: Graph Rewiring with Gumbel-Softmax |
| topic | Machine Learning |
| url | https://arxiv.org/abs/2508.17531 |