Discover the intricacies of Novel Class Discovery (NCD) and how it differs from related domains like Semi-Supervised Learning and Transfer Learning.

This conclusion summarizes insights on how topology awareness affects GNN generalization performance.

This case study demonstrates the application of shortest-path distance to evaluate GNN performance on structural subgroups.

Explore the main results on GNN generalization performance, highlighting how structural distance affects accuracy disparities across subgroups.

Explore the relationship between topology awareness and generalization performance in GNNs using metric distortion.

Review the nature of topology awareness in GNNs, including its impact on generalization, and active learning challenges.

Explore the influence of topology awareness on the generalization performance of Graph Neural Networks (GNNs) in this comprehensive study.

Discover a framework that enhances GNN learning, enabling fair k-shot learning and fairness constraint for equitable predictive performance in structural groups

This appendix details the experimental setup, hardware, hyperparameters, and additional results confirming GNN performance and findings across datasets.

Modified greedy and sampling algorithms solve Eq. (6) in GNNs for the k-center problem, with a running time complexity of O(k) and O(kn) respectively.

Learn how GNNs exhibit unfair generalization when comparing test groups with significant structural distance differences.