Gauge Geometry of Hodge Zero-Mode Transport in Parameter-Dependent Topological Data Analysis
Satoshi Kanno, Rei Nishimura, Hiroshi Yamauchi, Yoshi-aki Shimada · 2026
Read Gauge Geometry of Hodge Zero-Mode Transport in Parameter-Dependent Topological Data Analysis by Satoshi Kanno, Rei Nishimura, Hiroshi Yamauchi, Yoshi-aki Shimada (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2605-28326.txt.
Introduction
Kanno and colleagues follow shape traits as settings shift, past birth-death charts. They code traits as zero modes of the Hodge Laplacian in one shared chain space, then read bend for local mixing and loop-memory for cycle effects. On shifting point clouds the readouts spot wobble that point-wise matching misses.
Catalog record
| CREATOR | Satoshi Kanno, Rei Nishimura, Hiroshi Yamauchi, Yoshi-aki Shimada |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 120266 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2605.28326
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Gauge Geometry of Hodge Zero-Mode Transport in Parameter-Dependent Topological Data Analysis Satoshi Kanno, Rei Nishimura, Hiroshi Yamauchi, Yoshi-aki Shimada Abstract We propose a practical computational framework for detecting structural changes in parameter-dependent topological data. In many applications, such as time-series data analysis, anomaly detection, and monitoring of systems under changing control parameters, persistence diagrams describe the birth and death of topological features at each parameter value, but they do not fully capture how these features are reorganized over time. To address this limitation, we represent homological features by zero modes of the ordinary combinatorial Hodge Laplacian and track the corresponding feature spaces in a common ambient chain space. This allows us to compute curvature and holonomy as descriptors of local reorganization and accumulated memory in evolving topological structures. Curvature highlights parameter regions where homological features mix or change rapidly, while holonomy summarizes the net effect of such changes after a closed cycle. We also establish stability estimates showing that these descriptors are robust under perturbations of the Hodge Laplacian on regular regions. Numerical experiments on controlled time-dependent point-cloud data show that the proposed method detects tracking instability, distinguishes systems with nearly identical persistence diagrams, and captures cycle-level memory invisible to pointwise feature matching. These results suggest that zero-mode transport geometry can serve as a useful computational tool for analyzing dynamic topological data. keywords Topological data analysis, Hodge Laplacian, Parameter-dependent homology, Berry connection 1 Introduction Persistent homology 12 ; 37 ; 9 ; 11 ; 31 is one of the central tools in topological data analysis. It extracts topological structures from data through a filtration and summarizes their robustness in terms of birth and death. Barcodes and persistence diagrams provide concise representations of the lifetimes of topological features, and have been widely used in both theory and applications, including glassy states 30 ; 29 ; 20 , materials science, machine learning, and neural network analysis 27 ; 33 ; 13 ; 21 ; 1 ; 22 ; 4 ; 8 ; 23 . In many applications, data depend not only on a filtration parameter but also on an additional external parameter, such as time, an external field, a control variable, an environmental variable, or an experimental condition. In such situations, it is practically important to describe and quantify how persistent homology, or more generally homological features, change as the external parameter varies. For example, when filtrations are constructed at successive times from time-series medical images, large changes in homological structure may be used as indicators of abnormal behavior. Similarly, when time-series data are analyzed by delay-coordinate embedding, or Takens embedding, changing the observation window produces a family of filtrations depending on a window parameter. In this case, detecting changes in the resulting homological features is useful for anomaly detection and the analysis of state transitions. Thus, tracking homology along variations of an external parameter is not merely a mathematical problem, but an important practical task in data analysis. In particular, one needs tools that describe not only the birth and death of topological features at each parameter value, but also how the corresponding homological feature spaces themselves vary over parameter space. A representative viewpoint for such situations is given by vineyards 10 ; 7 . Vineyards track how points in persistence diagrams move as the underlying data vary, thereby visualizing the evolution of topological features. This viewpoint is intuitive and has the advantage of directly displaying the motion of features in the diagram. By contrast, zigzag persistence 5 extends interval-decomposition-type descriptions to diagrams with both forward and backward maps, rather than aiming to describe the transport of feature spaces along an external parameter. Likewise, much of multiparameter persistence treats the additional parameter itself as a filtration coordinate, so that the entire structure becomes a genuine multifiltration, as in the rank invariant 6 , fibered barcodes and their visualization in RIVET 24 , Hilbert functions 32 , and multiparameter landscapes 34 . The setting of the present work is different: the scale parameter is regarded as the filtration parameter, while the second variable is treated as an external parameter such as time or a control variable. Although vineyards provide an important description, any approach based on pointwise tracking has intrinsic limitations. When several significant persistence points approach one another, or when many short-lived features appear simultaneously, matching diagram points becomes unstable, and small perturbations can substantially change the correspondence. Moreover, even when local tracking is possible, it may fail to extend to a globally consistent labeling 19 ; 18 . Thus, parameter-dependent topological structure cannot always be fully described by trajectories of labeled points in persistence diagrams. This difficulty reflects the fact that homology is, at each parameter value, a finite-dimensional vector space. Choices of generators or representative cycles are generally not unique and need not be preserved under parameter variation. A basis that appears natural at one parameter value may become a different linear combination at another. Hence, tracking individual generators or feature labels is basis-dependent, whereas the intrinsic object is the variation of the homology space itself. In this work, we therefore shift attention from individual persistence-diagram points to the reorganization of homological feature spaces. Rather than asking only how diagram points move, we ask how the corresponding homology spaces are transported, mixed, and reorganized over parameter space. Since homology is a finite-dimensional vector space at each parameter value, these spaces can be regarded as fibers over parameter space, and their variation can be formulated as a transport problem for homological vector spaces. We make this transport geometry concrete using the ordinary combinatorial Hodge Laplacian. For a finite simplicial complex, the discrete Hodge theorem identifies the zero eigenspace of the Hodge Laplacian with homology. Thus, for a parameter-dependent filtration, the homology at each scale and external parameter value can be realized as a Hodge zero-mode space. The key point is that, instead of choosing individual generators or representative cycles, one can obtain a basis-independent description by using the orthogonal projection onto the zero-mode space. More concretely, we embed the zero-mode spaces into a common Hilbert space and regard them as a zero-mode bundle over regular regions where the zero-mode multiplicity is constant and a spectral gap is maintained. On this bundle, a change of basis acts as a gauge transformation, while the curvature defined by the projection formula, as well as spectral quantities, invariant norms, and trace-type quantities derived from holonomy, provide gauge-invariant information. Therefore, our method describes the reorganization of homological feature spaces through zero-mode projections and gauge-invariant geometric quantities, rather than through generator tracking that depends on choices of bases or labels. The proposed framework is not intended to replace vineyards. Rather, zero-mode transport is naturally connected to vineyard-type tracking. In regular regimes where persistence features are well separated and matching is stable, zero-mode transport is numerically consistent with vineyard-type monodromy. In this sense, the correspondences and exchanges described by vineyards can be viewed as coarse observations of transport in the zero-mode space. At the same time, because the proposed quantities are defined from the geometry of Hodge zero-mode subspaces rather than from a choice of labels or bases, transport, curvature, and holonomy remain stably computable even when persistence points approach one another and vineyard matching becomes ambiguous.