Test-time GNN Model Evaluation on Dynamic Graphs
This addresses performance uncertainty in deployed DGNNs for real-world dynamic systems, but it is incremental as it introduces a new evaluation method for an existing bottleneck.
The paper tackles the problem of evaluating dynamic graph neural networks (DGNNs) on unseen test graphs with distribution shifts, proposing DyGEval, a two-stage evaluator that estimates performance by simulating training-test differences, and shows it effectively assesses various DGNN backbones across dynamic graphs.
Dynamic graph neural networks (DGNNs) have emerged as a leading paradigm for learning from dynamic graphs, which are commonly used to model real-world systems and applications. However, due to the evolving nature of dynamic graph data distributions over time, well-trained DGNNs often face significant performance uncertainty when inferring on unseen and unlabeled test graphs in practical deployment. In this case, evaluating the performance of deployed DGNNs at test time is crucial to determine whether a well-trained DGNN is suited for inference on an unseen dynamic test graph. In this work, we introduce a new research problem: DGNN model evaluation, which aims to assess the performance of a specific DGNN model trained on observed dynamic graphs by estimating its performance on unseen dynamic graphs during test time. Specifically, we propose a Dynamic Graph neural network Evaluator, dubbed DyGEval, to address this new problem. The proposed DyGEval involves a two-stage framework: (1) test-time dynamic graph simulation, which captures the training-test distributional differences as supervision signals and trains an evaluator; and (2) DyGEval development and training, which accurately estimates the performance of the well-trained DGNN model on the test-time dynamic graphs. Extensive experiments demonstrate that the proposed DyGEval serves as an effective evaluator for assessing various DGNN backbones across different dynamic graphs under distribution shifts.