Computer Science > Machine Learning
[Submitted on 5 Feb 2024 (v1), last revised 7 Feb 2024 (this version, v2)]
Title:Statistical Guarantees for Link Prediction using Graph Neural Networks
Download PDFAbstract: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.
Submission history
From: Alan Chung [view email][v1] Mon, 5 Feb 2024 03:03:00 UTC (537 KB)
[v2] Wed, 7 Feb 2024 16:16:08 UTC (537 KB)
Current browse context:
cs.LG
References & Citations
Bibliographic and Citation Tools
Bibliographic Explorer (What is the Explorer?)
Litmaps (What is Litmaps?)
scite Smart Citations (What are Smart Citations?)