Geometric Imbalance in Semi-Supervised Node Classification
This addresses the problem of robust semi-supervised node classification for graph data practitioners, offering both theoretical insights and practical tools, though it appears incremental as it builds on existing imbalance mitigation concepts.
The paper tackles class imbalance in semi-supervised node classification by introducing geometric imbalance, which describes how message passing causes ambiguity for minority-class nodes in embedding spaces, and proposes a framework with pseudo-label alignment, node reordering, and ambiguity filtering to address it. Experiments show the approach consistently outperforms existing methods, particularly under severe imbalance.
Class imbalance in graph data presents a significant challenge for effective node classification, particularly in semi-supervised scenarios. In this work, we formally introduce the concept of geometric imbalance, which captures how message passing on class-imbalanced graphs leads to geometric ambiguity among minority-class nodes in the riemannian manifold embedding space. We provide a rigorous theoretical analysis of geometric imbalance on the riemannian manifold and propose a unified framework that explicitly mitigates it through pseudo-label alignment, node reordering, and ambiguity filtering. Extensive experiments on diverse benchmarks show that our approach consistently outperforms existing methods, especially under severe class imbalance. Our findings offer new theoretical insights and practical tools for robust semi-supervised node classification.