Discrimination of Dynamic Data via Curvature Sets

TL;DR

This paper introduces dynamic curvature-set persistent homology, a novel topological data analysis method for distinguishing time-dependent data, with improved computational efficiency and stability, demonstrated on the Boids model.

Contribution

We extend curvature-set persistent homology to dynamic data, proving interval- and antichain-decomposability of the resulting modules and developing an efficient algorithm for erosion distance computation.

Findings

Successfully distinguished parameter changes in the Boids model

Proved stability of the method under a generalized Gromov-Hausdorff distance

Established computational efficiency through antichain-decomposability

Abstract

Techniques from topological data analysis (TDA) have proven effective in studying time-dependent data arising in dynamic systems, such as animal swarming behavior and spatiotemporal patterns in neuroscience. While early algorithms leveraged efficient updates to persistence diagrams for dynamic data, they struggled to distinguish behaviors that are isometric at each fixed time but differ qualitatively. This limitation was addressed by Kim and M\'emoli, who introduced a spatiotemporal persistence framework for dynamic metric spaces, resulting in multiparameter persistence modules. However, these modules pose computational challenges. To address this, we build on insights from G\'omez and M\'emoli, who observed that the homology of Rips complexes over size $(2k+2)$ point subsets of a metric space--termed principal curvature sets--is both tractable and informative. We extend this idea to dynamic settings by introducing dynamic curvature-set persistent homology, applying the spatiotemporal framework of Kim and M\'emoli to curvature sets. We prove that the resulting multiparameter persistence modules are interval-decomposable: in fact, they possess a stronger property we term antichain-decomposable. Utilizing this property, we present a new algorithm to efficiently compute the erosion distance $d_E$ (due to Patel) between arbitrary antichain-decomposable modules (including, but not limited to modules produced by our construction). Additionally, our construction is stable with respect to a generalized Gromov-Hausdorff distance between time-dependent datasets proposed by Kim and M\'emoli. This enables a robust computational pipeline for distinguishing dynamic data, as demonstrated in experiments with the Boids model, where we successfully detect parameter changes.