pith. sign in

arxiv: 2605.15884 · v1 · pith:I7QUY4EXnew · submitted 2026-05-15 · ❄️ cond-mat.stat-mech

Cycle affinity and winding localize eigenvalues of Markov generators

Artemy Kolchinsky , Naruo Ohga , Sosuke Ito This is my paper

Pith reviewed 2026-05-19 19:18 UTC · model grok-4.3

classification ❄️ cond-mat.stat-mech
keywords Markov generatorsnonequilibrium cycleseigenvalue localizationcycle affinitywinding numberthermodynamic boundsrelaxation modesoscillatory dynamics
0
0 comments X

The pith

Each complex eigenvalue of a Markov generator is confined to a region set by the affinity of some nonequilibrium cycle and the winding number of its eigenvector.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper establishes that complex eigenvalues, which control oscillatory relaxation and linear response in Markov processes, are localized by the thermodynamic properties of cycles in the transition graph. It proves that the cycle affinity and the eigenvector's winding number together create a confining region for each such eigenvalue in the complex plane. This relation exposes a direct tradeoff in which stronger nonequilibrium driving restricts possible oscillation frequencies and decay rates. For systems with a single cycle the winding number equals the ordered index of the eigenvalue, which immediately supplies thermodynamic bounds on the slowest and fastest relaxing modes. In systems with multiple cycles the same construction unifies earlier inequalities and confirms the Uhl-Seifert ellipse conjecture.

Core claim

The central claim is that for any Markov generator, every complex eigenvalue is localized inside a region whose boundaries are fixed by the affinity of at least one nonequilibrium cycle together with the integer winding number of the associated eigenvector; in the unicyclic case this winding number coincides with the eigenvalue's position in the ordered spectrum, and the resulting bounds extend to multicyclic generators to prove the ellipse conjecture.

What carries the argument

nonequilibrium cycle affinity combined with eigenvector winding number, which together define a closed region in the complex plane that must contain the eigenvalue

If this is right

  • In unicyclic Markov generators the winding number equals the ordered eigenvalue index and therefore supplies explicit thermodynamic upper bounds on both the slowest and fastest relaxation rates.
  • The localization immediately yields new inequalities relating the imaginary part of any eigenvalue to the cycle affinity.
  • The same argument unifies and strengthens several previously known bounds on eigenvalues of multicyclic generators.
  • The construction proves the Uhl-Seifert ellipse conjecture for the location of eigenvalues in the complex plane.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • The localization could be used to design reaction networks or stochastic models whose dominant oscillation frequency is controlled by tuning a single cycle affinity.
  • Numerical checks on small multicyclic graphs would test whether the predicted regions remain tight when many overlapping cycles are present.
  • The result suggests that any violation of the localization would require either a cycle affinity of zero or a non-integer winding number, both of which are ruled out by standard Markov theory.

Load-bearing premise

For every complex eigenvalue there exists at least one nonequilibrium cycle whose affinity and the eigenvector's winding number together produce a confining region.

What would settle it

A concrete Markov generator whose spectrum contains a complex eigenvalue lying strictly outside every region constructed from its own nonequilibrium cycles and the winding numbers of the corresponding eigenvectors.

Figures

Figures reproduced from arXiv: 2605.15884 by Artemy Kolchinsky, Naruo Ohga, Sosuke Ito.

Figure 1
Figure 1. Figure 1: FIG. 1. Illustration of Theorem 1 for a unicyclic generator with [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: FIG. 2. (a) Theorem 4 illustrated on a 6-state generator with 3 [PITH_FULL_IMAGE:figures/full_fig_p005_2.png] view at source ↗
read the original abstract

The complex eigenvalues of Markov generators govern oscillatory properties of relaxation, autocorrelation, and linear response. Here we show that these eigenvalues are localized by nonequilibrium cycles of the generator, thus revealing a fundamental tradeoff between thermodynamic driving, oscillation, and decay of eigenmodes. Specifically, we prove that each complex eigenvalue is confined to a region determined by the cycle affinity and the eigenvector ``winding number'' of some nonequilibrium cycle. In unicyclic systems, we also demonstrate that the winding number coincides with the ordered eigenvalue index, yielding new thermodynamic bounds on the slowest and fastest relaxation modes. In multicyclic systems, our approach unifies and extends several previous inequalities and proves the Uhl--Seifert ellipse conjecture.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

2 major / 2 minor

Summary. The manuscript proves that complex eigenvalues of continuous-time Markov generators are localized inside regions fixed by the affinity A_C and eigenvector winding number w_C of some nonequilibrium cycle C. In unicyclic networks the winding number equals the ordered index of the eigenvalue, supplying thermodynamic bounds on the slowest and fastest modes. For multicyclic generators the same construction unifies earlier inequalities and establishes the Uhl-Seifert ellipse conjecture.

Significance. If the central localization theorem holds, the work supplies a direct link between thermodynamic driving, oscillatory relaxation, and spectral decay that is absent from standard Perron-Frobenius or cycle-decomposition treatments. The mathematical proofs of the localization bound and of the ellipse conjecture constitute the principal strengths; they are parameter-free once the generator is given and yield falsifiable predictions for nonequilibrium Markov processes.

major comments (2)
  1. [Main theorem] Main theorem (multicyclic case): the claim that every complex eigenvalue λ is confined by the affinity and winding number of some nonequilibrium cycle C requires a constructive guarantee that such a cycle exists for an arbitrary eigenvector. Standard cycle decomposition does not automatically furnish one for every mode; the proof must therefore exhibit an explicit selection rule or show that the bound remains valid even if the cycle is chosen from the support of the eigenvector.
  2. [Unicyclic systems] § on unicyclic systems: the identification of the winding number with the ordered eigenvalue index is used to bound both the slowest and fastest relaxation rates. The derivation should be checked for hidden assumptions on the ordering of the real parts when the affinity is fixed.
minor comments (2)
  1. [Notation] Notation for the winding number w_C should be introduced with an explicit definition (phase accumulation along the cycle) before its first use in the localization statement.
  2. [Figures] Figure captions for the ellipse conjecture illustrations should state the numerical values of the affinities used so that the plotted boundaries can be reproduced from the generator alone.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the positive evaluation and for the detailed comments that help improve the clarity of the main results. We address each major comment below.

read point-by-point responses

  1. Referee: [Main theorem] Main theorem (multicyclic case): the claim that every complex eigenvalue λ is confined by the affinity and winding number of some nonequilibrium cycle C requires a constructive guarantee that such a cycle exists for an arbitrary eigenvector. Standard cycle decomposition does not automatically furnish one for every mode; the proof must therefore exhibit an explicit selection rule or show that the bound remains valid even if the cycle is chosen from the support of the eigenvector.

    Authors: We thank the referee for raising this important point on constructivity. In the proof of the localization theorem, the cycle C is selected constructively as any simple cycle lying in the support of the eigenvector (i.e., using only edges where the corresponding components of the right eigenvector are nonzero). The winding number w_C is then the integer winding of the argument of the eigenvector components around this cycle, which is guaranteed to exist by the fact that the eigenvector cannot be supported on a tree (otherwise it would be real). This selection is made explicit in the argument following Eq. (12) and does not rely on a global cycle decomposition of the entire graph. We will add a short paragraph in the revised manuscript that states this selection rule explicitly and notes that the resulting bound holds for any such cycle in the eigenvector support. revision: partial

  2. Referee: [Unicyclic systems] § on unicyclic systems: the identification of the winding number with the ordered eigenvalue index is used to bound both the slowest and fastest relaxation rates. The derivation should be checked for hidden assumptions on the ordering of the real parts when the affinity is fixed.

    Authors: We have carefully rechecked the derivation in the unicyclic section. The identification of the winding number w with the ordered index k of the eigenvalue λ_k proceeds from the explicit solution of the characteristic equation for the cycle graph and does not presuppose any ordering of the real parts beyond the standard fact that the Perron eigenvalue is real and simple. For fixed affinity A, the eigenvalues lie on the boundary of the region defined by the affinity and winding; the slowest mode corresponds to the smallest |Re(λ)| among those with w=1 and the fastest to the largest |Re(λ)| among admissible windings. No additional assumption on the ordering of Re(λ) for different windings is used. We will insert a brief clarifying sentence confirming this point. revision: partial

Circularity Check

0 steps flagged

Derivation is self-contained; no circular reductions identified

full rationale

The paper advances a mathematical localization theorem for complex eigenvalues of Markov generators, expressing bounds in terms of cycle affinity A_C and eigenvector winding number w_C, both defined directly from the generator matrix and its eigenvectors. These are not fitted parameters, self-referential definitions, or quantities obtained by renaming known results. The proof invokes standard cycle decomposition and Perron-Frobenius properties without load-bearing self-citations that reduce the central claim to prior unverified work by the same authors. The existence of a suitable cycle for each eigenvalue is an explicit assumption in the theorem rather than a hidden tautology, and the derivation does not collapse any prediction or bound to its own inputs by construction. The result is therefore independent of the inputs it localizes.

Axiom & Free-Parameter Ledger

0 free parameters · 2 axioms · 1 invented entities

The analysis rests on the standard framework of continuous-time Markov chains and introduces the winding-number concept to localize eigenvalues; no free parameters are fitted and no new physical entities are postulated beyond the mathematical definition of winding.

axioms (2)
  • domain assumption The dynamics are governed by a continuous-time Markov generator whose spectrum controls relaxation and response.
    This is the foundational modeling choice for the entire spectral analysis.
  • domain assumption Nonequilibrium cycles exist in the state space and possess well-defined affinities.
    Cycle affinity is the central quantity used to bound the eigenvalues.
invented entities (1)
  • eigenvector winding number no independent evidence
    purpose: Quantifies the phase accumulation of an eigenmode around a nonequilibrium cycle for localization purposes.
    The winding number is defined within the paper to extend spectral bounds beyond previous work.

pith-pipeline@v0.9.0 · 5645 in / 1428 out tokens · 68335 ms · 2026-05-19T19:18:04.811778+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Lean theorems connected to this paper

Citations machine-checked in the Pith Canon. Every link opens the source theorem in the public Lean library.

What do these tags mean?
matches
The paper's claim is directly supported by a theorem in the formal canon.
supports
The theorem supports part of the paper's argument, but the paper may add assumptions or extra steps.
extends
The paper goes beyond the formal theorem; the theorem is a base layer rather than the whole result.
uses
The paper appears to rely on the theorem as machinery.
contradicts
The paper's claim conflicts with a theorem or certificate in the canon.
unclear
Pith found a possible connection, but the passage is too broad, indirect, or ambiguous to say the theorem truly supports the claim.

Reference graph

Works this paper leans on

48 extracted references · 48 canonical work pages

  1. [1]

    1 (Elsevier, 1992)

    N.G.VanKampen,Stochasticprocessesinphysicsandchemistry, Vol. 1 (Elsevier, 1992)

    work page 1992

  2. [2]

    Schnakenberg, Network theory of microscopic and macro- scopic behavior of master equation systems, Reviews of Modern physics48, 571 (1976)

    J. Schnakenberg, Network theory of microscopic and macro- scopic behavior of master equation systems, Reviews of Modern physics48, 571 (1976)

    work page 1976

  3. [3]

    Qian, Thermodynamic and kinetic analysis of sensitivity am- plification in biological signal transduction, Biophysical chem- istry105, 585 (2003)

    H. Qian, Thermodynamic and kinetic analysis of sensitivity am- plification in biological signal transduction, Biophysical chem- istry105, 585 (2003)

    work page 2003

  4. [4]

    Skoge, S

    M. Skoge, S. Naqvi, Y. Meir, and N. S. Wingreen, Chemical sensingbynonequilibriumcooperativereceptors,Physicalreview letters110, 248102 (2013)

    work page 2013

  5. [5]

    Mehta and D

    P. Mehta and D. J. Schwab, Energetic costs of cellular compu- tation, Proceedings of the National Academy of Sciences109, 17978 (2012)

    work page 2012

  6. [6]

    L. J. Allen,An introduction to stochastic processes with applica- tions to biology(CRC press, 2010)

    work page 2010

  7. [7]

    T. M. Liggett,Stochastic interacting systems: contact, voter and exclusion processes, Vol. 324 (springer science & Business Media, 2013)

    work page 2013

  8. [8]

    A. C. Barato and U. Seifert, Coherence of biochemical oscilla- tionsisboundedbydrivingforceandnetworktopology,Physical Review E95, 062409 (2017)

    work page 2017

  9. [9]

    del Junco and S

    C. del Junco and S. Vaikuntanathan, High chemical affinity increases the robustness of biochemical oscillations, Physical Review E101, 012410 (2020)

    work page 2020

  10. [10]

    Oberreiter, U

    L. Oberreiter, U. Seifert, and A. C. Barato, Universal minimal cost of coherent biochemical oscillations, Physical Review E 106, 014106 (2022)

    work page 2022

  11. [11]

    Zheng and E

    C. Zheng and E. Tang, A topological mechanism for robust and efficient global oscillations in biological networks, Nature Communications15, 6453 (2024)

    work page 2024

  12. [12]

    M.UhlandU.Seifert,Affinity-dependentboundonthespectrum of stochastic matrices, Journal of Physics A: Mathematical and Theoretical52, 405002 (2019)

    work page 2019

  13. [13]

    N.Shiraishi,Entropyproductionlimitsallfluctuationoscillations, Physical Review E108, L042103 (2023)

    work page 2023

  14. [14]

    Kolchinsky, N

    A. Kolchinsky, N. Ohga, and S. Ito, Thermodynamic bound on spectral perturbations, with applications to oscillations and re- laxation dynamics, Physical Review Research6, 013082 (2024)

    work page 2024

  15. [15]

    A.Y.Mitrophanov,Thespectralgapandperturbationboundsfor reversible continuous-time Markov chains, Journal of applied probability41, 1219 (2004)

    work page 2004

  16. [16]

    B.GaveauandL.Schulman,Theoryofnonequilibriumfirst-order phase transitions for stochastic dynamics, Journal of Mathemati- cal Physics39, 1517 (1998)

    work page 1998

  17. [17]

    B.GaveauandL.Schulman,Multiplephasesinstochasticdynam- ics: Geometry and probabilities, Physical Review E—Statistical, Nonlinear, and Soft Matter Physics73, 036124 (2006)

    work page 2006

  18. [18]

    Raz,Eigenvaluecrossingasa phase transition in relaxation dynamics, Physical Review Letters130, 207103 (2023)

    G.Teza, R.Yaacoby,andO. Raz,Eigenvaluecrossingasa phase transition in relaxation dynamics, Physical Review Letters130, 207103 (2023)

    work page 2023

  19. [19]

    Seifert, Stochastic thermodynamics, fluctuation theorems and molecular machines, Reports on Progress in Physics75, 126001 (2012)

    U. Seifert, Stochastic thermodynamics, fluctuation theorems and molecular machines, Reports on Progress in Physics75, 126001 (2012). 7

    work page 2012

  20. [20]

    Wei and C

    J. Wei and C. D. Prater, The Structure and Analysis of Complex Reaction Systems, inAdvances in Catalysis, Vol. 13 (Elsevier,

    work page

  21. [21]

    J.Z.Hearon,Thekineticsoflinearsystemswithspecialreference to periodic reactions, The bulletin of mathematical biophysics 15, 121 (1953)

    work page 1953

  22. [22]

    N. Ohga, S. Ito, and A. Kolchinsky, Thermodynamic bound on the asymmetry of cross-correlations, Physical Review Letters 131, 077101 (2023)

    work page 2023

  23. [23]

    Nguyen, U

    B. Nguyen, U. Seifert, and A. C. Barato, Phase transition in ther- modynamically consistent biochemical oscillators, The Journal of Chemical Physics149, 045101 (2018)

    work page 2018

  24. [24]

    Remlein, B

    J. Remlein, B. Weissmann, and U. Seifert, Coherence of oscil- lations in the weak-noise limit, Physical Review E105, 064101 (2022)

    work page 2022

  25. [25]

    T.VanVu,V.T.Vo,andK.Saito,Dissipation,quantumcoherence, andasymmetryoffinite-timecross-correlations,PhysicalReview Research6, 013273 (2024)

    work page 2024

  26. [26]

    G.-H. Xu, A. Kolchinsky, J.-C. Delvenne, and S. Ito, Thermo- dynamic geometric constraint on the spectrum of markov rate matrices, Physical Review Letters135, 257102 (2025)

    work page 2025

  27. [27]

    N. M. Dmitriev and E. B. Dynkin, On the characteristic roots of stochastic matrices, Izvestiya Akademii Nauk SSSR. Seriya Matematicheskaya10,167(1946),englishtranslationofRussian title

    work page 1946

  28. [28]

    Kellogg and A

    R. Kellogg and A. Stephens, Complex eigenvalues of a non- negative matrix with a specified graph, Linear Algebra and its Applications20, 179 (1978)

    work page 1978

  29. [29]

    Wierenga, P

    H. Wierenga, P. R. Ten Wolde, and N. B. Becker, Quantifying fluctuations in reversible enzymatic cycles and clocks, Physical Review E97, 042404 (2018)

    work page 2018

  30. [30]

    Wachtel, R

    A. Wachtel, R. Rao, and M. Esposito, Thermodynamically con- sistent coarse graining of biocatalysts beyond michaelis–menten, New Journal of Physics20, 042002 (2018)

    work page 2018

  31. [31]

    A. B. Kolomeisky and M. E. Fisher, Molecular motors: a theorist’s perspective, Annual Review of Physical Chemistry58, 675 (2007)

    work page 2007

  32. [32]

    Seifert, Efficiency of autonomous soft nanomachines at maxi- mum power, Physical review letters106, 020601 (2011)

    U. Seifert, Efficiency of autonomous soft nanomachines at maxi- mum power, Physical review letters106, 020601 (2011)

    work page 2011

  33. [33]

    A.Piñero-SoléandA.Kolchinsky,Lightspectralresponsesofthe bacteriorhodopsin proton pump reveal directional and reciprocal interactions, Physical Review Research6, 013275 (2024)

    work page 2024

  34. [34]

    Maes, Local detailed balance, SciPost Phys

    C. Maes, Local detailed balance, SciPost Phys. Lect. Notes , 32 (2021)

    work page 2021

  35. [35]

    R. S. Varga,Geršgorin and His Circles, Springer Series in Com- putational Mathematics, Vol. 36 (Springer Berlin Heidelberg, Berlin, Heidelberg, 2004)

    work page 2004

  36. [36]

    See supplemental material for additional proofs and derivations

    work page

  37. [37]

    Braverman, R

    M.Shubin,InvitationtoPartialDifferentialEquations,editedby M. Braverman, R. McOwen, and P. Topalov, Graduate Studies in Mathematics, Vol. 205 (American Mathematical Society, Providence, Rhode Island, 2020)

    work page 2020

  38. [38]

    Bohner, O

    M. Bohner, O. Došlý, and W. Kratz, An Oscillation Theorem for Discrete Eigenvalue Problems, The Rocky Mountain Journal of Mathematics33, 1233 (2003)

    work page 2003

  39. [39]

    Reimann, Brownian motors: noisy transport far from equilib- rium, Physics Reports361, 57 (2002)

    P. Reimann, Brownian motors: noisy transport far from equilib- rium, Physics Reports361, 57 (2002)

    work page 2002

  40. [40]

    F. I. Karpelevich, On the characteristic roots of matrices with non-negative elements, Izvestiya Akademii Nauk SSSR. Seriya Matematicheskaya15, 361 (1951)

    work page 1951

  41. [41]

    Dasdan and R

    A. Dasdan and R. K. Gupta, Faster maximum and minimum mean cycle algorithms for system-performance analysis, IEEE transactions on computer-aided design of integrated circuits and systems17, 889 (1998)

    work page 1998

  42. [42]

    T.BenAvraham,G.Sharon,Y.Zarai,andM.Margaliot,Dynam- ical systems with a cyclic sign variation diminishing property, IEEE Transactions on Automatic Control65, 941 (2020)

    work page 2020

  43. [43]

    crossings

    R.Alseidi,M.Margaliot,andJ.Garloff,Onthespectralproperties of nonsingular matrices that are strictly sign-regular for some order with applications to totally positive discrete-time systems, Journal of Mathematical Analysis and Applications474, 524 (2019). 8 SUPPLEMENTAL MATERIAL SM 1: UNICYCLIC RESULTS We show that the unicyclic winding number can be equi...

    work page 2019

  44. [44]

    It is straightforward to verify that the winding number(8) is invariant to conjugation, so ωc(u) =ω c(u∗) for any admissible cycle where they are well- defined

    To show the result for a nonreal left eigenpair(λ,u) with Imλ <0 , one may consider the conjugate eigenpair(λ∗,u ∗) with Imλ ∗ =−Imλ >0 . It is straightforward to verify that the winding number(8) is invariant to conjugation, so ωc(u) =ω c(u∗) for any admissible cycle where they are well- defined. Therefore, if the theorem holds for eigenpairs with Imλ >0...

    work page

  45. [45]

    The vertex set is non-empty:V̸=∅

    work page

  46. [46]

    Every vertex i∈V has nonnegative total outgoing weight:P j∈V:(j←i)∈E fj←i ≥0

    work page

  47. [47]

    Every vertexi∈V has an outgoing edge(j←i)∈E to somej∈V

    work page

  48. [48]

    roundtrips

    Every nonnegative reversible edge has negative reverse weight: if (j←i)∈E , fj←i ≥0 , and(i←j)∈E , thenf i←j <0. Then, G contains a directed simple cyclec that satisfies one of the following: A. All edges(j←i)∈c are nonnegative (fj←i ≥0 ) and at least one is irreversible,(i←j)/∈E; B. All edges (j←i)∈c are positive and reversible (fj←i >0,(i←j)∈E) and sati...

    work page