Topology-Aware Active Learning on Graphs
This work addresses the problem of efficient label acquisition for graph classification tasks, offering a domain-specific improvement over existing methods.
The paper tackles the challenge of exploration versus exploitation in graph-based active learning under scarce label budgets by introducing a coreset construction algorithm using Balanced Forman Curvature and a localized graph rewiring strategy, resulting in consistent outperformance of existing baselines at low label rates.
We propose a graph-topological approach to active learning that directly targets the core challenge of exploration versus exploitation under scarce label budgets. To guide exploration, we introduce a coreset construction algorithm based on Balanced Forman Curvature (BFC), which selects representative initial labels that reflect the graph's cluster structure. This method includes a data-driven stopping criterion that signals when the graph has been sufficiently explored. We further use BFC to dynamically trigger the shift from exploration to exploitation within active learning routines, replacing hand-tuned heuristics. To improve exploitation, we introduce a localized graph rewiring strategy that efficiently incorporates multiscale information around labeled nodes, enhancing label propagation while preserving sparsity. Experiments on benchmark classification tasks show that our methods consistently outperform existing graph-based semi-supervised baselines at low label rates.