Quantifying the structural stability of simplicial homology
Nicola Guglielmi, Anton Savostianov, Francesco Tudisco
TL;DR
The paper addresses the problem of quantifying the structural stability of a simplicial complex by identifying the smallest edge perturbation that increases the dimension of the $1$-st homology group. It casts this as a spectral matrix nearness problem using weighted, normalized higher-order Laplacians and tackles it with a bilevel optimization: a constrained-gradient inner loop minimizes a spectral functional $F(\varepsilon,E)$, while an outer level adjusts the perturbation size $\varepsilon$ to drive $F$ to zero. A key contribution is the functional $F(\varepsilon,E) = \tfrac{1}{2} \lambda_+(\varepsilon,E)^2 + \tfrac{\alpha}{2}\max\{0,1-\mu_2(\varepsilon,E)/\mu\}^2$, which penalizes near-disconnectedness while driving the up-Laplacian eigenvalue to zero, thereby inducing a new hole with minimal perturbation. The framework is validated on synthetic quasi-triangulations and real transportation networks, illustrating the method’s ability to quantify topological instability and guiding edge-pruning decisions in higher-order relational data.
Abstract
The homology groups of a simplicial complex reveal fundamental properties of the topology of the data or the system and the notion of topological stability naturally poses an important yet not fully investigated question. In the current work, we study the stability in terms of the smallest perturbation sufficient to change the dimensionality of the corresponding homology group. Such definition requires an appropriate weighting and normalizing procedure for the boundary operators acting on the Hodge algebra's homology groups. Using the resulting boundary operators, we then formulate the question of structural stability as a spectral matrix nearness problem for the corresponding higher-order graph Laplacian. We develop a bilevel optimization procedure suitable for the formulated matrix nearness problem and illustrate the method's performance on a variety of synthetic quasi-triangulation datasets and transportation networks.
