Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Preprint |
| Published: |
2024
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2408.16871 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866929479181402112 |
|---|---|
| author | Rasti-Meymandi, Arash Sajedi, Ahmad Xu, Zhaopan Plataniotis, Konstantinos N. |
| author_facet | Rasti-Meymandi, Arash Sajedi, Ahmad Xu, Zhaopan Plataniotis, Konstantinos N. |
| contents | Graph distillation has emerged as a solution for reducing large graph datasets to smaller, more manageable, and informative ones. Existing methods primarily target node classification, involve computationally intensive processes, and fail to capture the true distribution of the full graph dataset. To address these issues, we introduce Graph Distillation with Structural Attention Matching (GSTAM), a novel method for condensing graph classification datasets. GSTAM leverages the attention maps of GNNs to distill structural information from the original dataset into synthetic graphs. The structural attention-matching mechanism exploits the areas of the input graph that GNNs prioritize for classification, effectively distilling such information into the synthetic graphs and improving overall distillation performance. Comprehensive experiments demonstrate GSTAM's superiority over existing methods, achieving 0.45% to 6.5% better performance in extreme condensation ratios, highlighting its potential use in advancing distillation for graph classification tasks (Code available at https://github.com/arashrasti96/GSTAM). |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2408_16871 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | GSTAM: Efficient Graph Distillation with Structural Attention-Matching Rasti-Meymandi, Arash Sajedi, Ahmad Xu, Zhaopan Plataniotis, Konstantinos N. Machine Learning Artificial Intelligence Graph distillation has emerged as a solution for reducing large graph datasets to smaller, more manageable, and informative ones. Existing methods primarily target node classification, involve computationally intensive processes, and fail to capture the true distribution of the full graph dataset. To address these issues, we introduce Graph Distillation with Structural Attention Matching (GSTAM), a novel method for condensing graph classification datasets. GSTAM leverages the attention maps of GNNs to distill structural information from the original dataset into synthetic graphs. The structural attention-matching mechanism exploits the areas of the input graph that GNNs prioritize for classification, effectively distilling such information into the synthetic graphs and improving overall distillation performance. Comprehensive experiments demonstrate GSTAM's superiority over existing methods, achieving 0.45% to 6.5% better performance in extreme condensation ratios, highlighting its potential use in advancing distillation for graph classification tasks (Code available at https://github.com/arashrasti96/GSTAM). |
| title | GSTAM: Efficient Graph Distillation with Structural Attention-Matching |
| topic | Machine Learning Artificial Intelligence |
| url | https://arxiv.org/abs/2408.16871 |