Online GNN Evaluation Under Test-time Graph Distribution Shifts

arXiv:2403.09953v1 Announce Type: new
Abstract: Evaluating the performance of a well-trained GNN model on real-world graphs is a pivotal step for reliable GNN online deployment and serving. Due to a lack of test node labels and unknown potential training-test graph data distribution shifts, conventional model evaluation encounters limitations in calculating performance metrics (e.g., test error) and measuring graph data-level discrepancies, particularly when the training graph used for developing GNNs remains unobserved during test time. In this paper, we study a …

abstract arxiv cs.lg data deployment distribution evaluation gnn graph graph data graphs labels limitations metrics node performance pivotal test training type world