Statistical Guarantees for Link Prediction using Graph Neural Networks
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , |
|---|---|
| Format: | Preprint |
| Publié: |
2024
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866916117790851072 |
|---|---|
| author | Chung, Alan Saberi, Amin Austern, Morgane |
| author_facet | Chung, Alan Saberi, Amin Austern, Morgane |
| contents | This paper derives statistical guarantees for the performance of Graph Neural Networks (GNNs) in link prediction tasks on graphs generated by a graphon. We propose a linear GNN architecture (LG-GNN) that produces consistent estimators for the underlying edge probabilities. We establish a bound on the mean squared error and give guarantees on the ability of LG-GNN to detect high-probability edges. Our guarantees hold for both sparse and dense graphs. Finally, we demonstrate some of the shortcomings of the classical GCN architecture, as well as verify our results on real and synthetic datasets. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2402_02692 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Statistical Guarantees for Link Prediction using Graph Neural Networks Chung, Alan Saberi, Amin Austern, Morgane Machine Learning Social and Information Networks Statistics Theory This paper derives statistical guarantees for the performance of Graph Neural Networks (GNNs) in link prediction tasks on graphs generated by a graphon. We propose a linear GNN architecture (LG-GNN) that produces consistent estimators for the underlying edge probabilities. We establish a bound on the mean squared error and give guarantees on the ability of LG-GNN to detect high-probability edges. Our guarantees hold for both sparse and dense graphs. Finally, we demonstrate some of the shortcomings of the classical GCN architecture, as well as verify our results on real and synthetic datasets. |
| title | Statistical Guarantees for Link Prediction using Graph Neural Networks |
| topic | Machine Learning Social and Information Networks Statistics Theory |
| url | https://arxiv.org/abs/2402.02692 |