arXiv Science⌕ Search

arXiv · 2611.09968

Topology and Information in Regularized Transport on Graphs

Abstract

We study how two geometries of mass on a finite graph interact: the information geometry of the generalized entropy $F_q$ used for regularized transport, and the arithmetic of divisors governed by the Riemann--Roch theorem of Baker and Norine and by the Jacobian. Because the correction term of $F_q$ is affine, $F_q$-regularization at strength $γ$ is Tsallis regularization at temperature $γ/q$. Over the reals, chip-firing connects all configurations of equal degree, so the Jacobian enters transport only through integrality. We introduce the transport defect, the worst-case cost of steering $d$ chips towards a target by chip-firing, and prove $1/d\leδ_{\TV}(D)\le(d-r(D))/d=(g-1-r(K-D))/d\le\min(1,g/d)$ whenever $g\ge1$; vertex multiplicities give a lower bound and an equality criterion. A regularized topological gap $Γ_{q,γ}(D,E)$ tends to the chip-move defect as $γ\to0$, lies in an explicit window, has a high-temperature slope given by Tsallis coupling entropies, and admits a positive regime if and only if $g\ge1$. Dual potentials certify exactly the convex relaxation of a linear system; the remainder is an integrality gap. Zeros in optimal plans occur only for $q>1$, by a mechanism different from the sign change of $Γ$. Exact computations on $1784$ divisor classes find the rank bound $(d-r(D))/d$ attained in $1769$ of them, an observed frequency. Experiments on honeycomb and Erdős--Rényi networks with $10^2$ to $10^5$ nodes test the statements on regularized plans, not $Γ$; over $10^3\le n\le10^5$ their transport costs are consistent with growth like $n^{1/2}$ and $\log n$, respectively.

Explore related subjects

Keep this discovery

Explore connections, maps & timelines

BibTeXRIS

Shintaro Yoshizawa. 2026-10-30. Topology and Information in Regularized Transport on Graphs. https://arxiv.org/abs/2611.09968

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related papers

Broadband molecular dynamics simulation of confined Brownian motion

Molecular dynamics with an explicit solvent resolves the hydrodynamic response of a Brownian colloid in bulk and near a rigid wall over acoustic, inertial, and diffusive time scales. Bulk spectra establish a continuum reference and reveal departures associated with discrete-solvent dynamics. Under confinement, thermal spectra provide a measurement of the space-dependent added mass, while velocity autocorrelations determine space-dependent, anisotropic hydrodynamic memory and diffusion. All together, these measurements connect the enhanced added mass and reduced mobility near a wall, and probe the discrete-continuum crossover.

cond-mat.soft↗

Tunable Signal Penetration and Response Plateaus in Bistable Mechanical Media

Dynamic processing of mechanical signals is crucial for soft robotics and mechanosensing, yet classical viscoelastic materials lack intrinsic tunability and quantitative design rules. We show that internal bistability provides a simple but powerful mechanism to tailor response and signal attenuation in mechanical (meta)materials. In our model, bistable elements switch discretely with an intrinsic timescale between states that differ in potential energy $ε$, equilibrium length $Δl$, and spring constant $Δk$. We combine Brownian dynamics simulations with discrete Poisson switching at rate $ν$ and a nonlinear continuum (field) theory for the macroscopic strain relaxation. The model yields closed-form solutions for the linear response and the spatial penetration depth, revealing two complementary phenomena: a universal screening mechanism (akin to the electrostatic `skin effect') that strongly reduces signal penetration when the driving frequency exceeds the internal relaxation rate, and a frequency-insensitive response plateau arising from timescale separation. The attenuation strength, expressed by a screening length, is controlled primarily by the conformational length change $Δl$, while the attenuation regime and plateau position are tuneable via $ν$. A parameter study exposes a fundamental design trade-off: larger $Δl$ strengthen dissipation but also raise the elastic energy barrier for state transitions, eventually causing state-locking where damping vanishes. Optimal attenuation thus emerges from a compromise between pronounced bistability and a surmountable barrier. Owing to its analytical tractability and generic formulation, our framework yields explicit design rules applicable to systems from biopolymers to synthetic catch bonds and metamaterials, enabling the design of soft materials with frequency-selective mechanical response.

cond-mat.soft↗

A numerical study to analyze the interplay of Weissenberg number and viscosity ratio in a log-strain tensorial model for viscoelastic fluids

We present a computational study aimed at exploring the different and independent roles of the Weissenberg number and of the ratio between the polymeric and solvent viscosity contributions in a viscoelastic fluid model. The tensorial model under investigation, recently proposed, is based on a logarithmic relation between the elastic (or recoverable) strain and the elastic stress. In this model, the elastic strain plays the role of a conformation tensor and its evolution equation inherently preserves its determinant and positive definiteness. These properties are also enforced in the computational method employed in the study. A finite-difference discretization in time is combined with a stabilized mixed finite element formulation based on the Variational Multiscale method for the spatial discretization and with a generalized Lie derivative approach for the advection terms. The behavior of the model is analyzed in paradigmatic pressure-driven flows and we find that the value of the viscosity ratio is crucial in determining to which extent non-Newtonian flow profiles are observed upon increasing the Weissenberg number. By comparing the solutions of the log-strain tensorial model with those of a suitable Generalized Newtonian Fluid model, we show that flow-type dependence and memory play a significant role even in the simple planar flow past a cylinder.

cond-mat.soft↗