Markov and lattice bases for Forman-Ricci curvature of graphs

arXiv:2608.01929 2026 Regularization 1 ideas extracted · analyzed Aug 31, 2026

What the math gives to ML

The paper develops integer-preserving local rewiring moves for traversing graph fibers with fixed vertex-degree and Forman-curvature statistics. The transferable asset is not curvature itself, but the construction of small, exactly constraint-preserving perturbations: a graph can be augmented while retaining prescribed structural marginals, while degree-three lattice moves provide a cheap move set even when indispensable Markov moves are large. This suggests a controlled graph-neural-network augmentation and topology-robustness method in which training views differ in connectivity but remain matched on degree and curvature distributions.

Ideas from this paper

Unverified 2026

Curvature-Preserving Graph Augmentation

Generate alternative graph views by applying small integer Markov moves to the joint degree matrix, while rejecting moves that violate nonnegativity or realizability as a simple graph. Train a GNN to produce consistent predictions across the original and rewired views, preserving degree frequencies and curvature-frequency statistics while forcing robustness to higher-order topology.

Useful6/10
Difficulty6/10
Novelty7/10
Paper: Markov and lattice bases for Forman-Ricci curvature of graphs arXiv:2608.01929