Adaptive Graph Coarsening for Efficient GNN Training
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_ | 1866918151432699904 |
|---|---|
| author | Olshevskyi, Rostyslav Navarro, Madeline Segarra, Santiago |
| author_facet | Olshevskyi, Rostyslav Navarro, Madeline Segarra, Santiago |
| contents | We propose an adaptive graph coarsening method to jointly learn graph neural network (GNN) parameters and merge nodes via K-means clustering during training. As real-world graphs grow larger, processing them directly becomes increasingly challenging and sometimes infeasible. Tailoring algorithms to large-scale data may sacrifice performance, so we instead consider graph reduction to decrease the amount of data used during training. In particular, we propose a method to simultaneously train a GNN and coarsen its graph by partitioning nodes via K-means clustering based on their embeddings. Unlike past graph coarsening works, our approach allows us to merge nodes during training. Not only does this preclude coarsening as a preprocessing step, but our node clusters can adapt to the learning task instead of relying solely on graph connectivity and features. Thus, our method is amenable to scenarios that are challenging for other methods, such as heterophilic data. We validate our approach on both homophilic and heterophilic node classification datasets. We further visualize relationships between node embeddings and their corresponding clusters to illustrate that our coarsened graph adapts to the learning task during training. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2509_25706 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Adaptive Graph Coarsening for Efficient GNN Training Olshevskyi, Rostyslav Navarro, Madeline Segarra, Santiago Machine Learning We propose an adaptive graph coarsening method to jointly learn graph neural network (GNN) parameters and merge nodes via K-means clustering during training. As real-world graphs grow larger, processing them directly becomes increasingly challenging and sometimes infeasible. Tailoring algorithms to large-scale data may sacrifice performance, so we instead consider graph reduction to decrease the amount of data used during training. In particular, we propose a method to simultaneously train a GNN and coarsen its graph by partitioning nodes via K-means clustering based on their embeddings. Unlike past graph coarsening works, our approach allows us to merge nodes during training. Not only does this preclude coarsening as a preprocessing step, but our node clusters can adapt to the learning task instead of relying solely on graph connectivity and features. Thus, our method is amenable to scenarios that are challenging for other methods, such as heterophilic data. We validate our approach on both homophilic and heterophilic node classification datasets. We further visualize relationships between node embeddings and their corresponding clusters to illustrate that our coarsened graph adapts to the learning task during training. |
| title | Adaptive Graph Coarsening for Efficient GNN Training |
| topic | Machine Learning |
| url | https://arxiv.org/abs/2509.25706 |