pith. sign in

arxiv: 2011.00690 · v2 · submitted 2020-11-02 · 🌊 nlin.SI · math-ph· math.MP· math.PR

Bi-infinite solutions for KdV- and Toda-type discrete integrable systems based on path encodings

David A. Croydon , Makiko Sasada , Satoshi Tsujimoto This is my paper

Pith reviewed 2026-05-24 14:37 UTC · model grok-4.3

classification 🌊 nlin.SI math-phmath.MPmath.PR
keywords bi-infinite solutionspath encodingsdiscrete KdV equationdiscrete Toda equationultra-discrete equationsPitman's transformationcarrier processintegrable systems
0
0 comments X

The pith

Path encodings yield unique solutions to the initial value problem for bi-infinite discrete KdV and Toda equations when data lies in a suitable class.

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

The paper defines bi-infinite versions of the ultra-discrete KdV, discrete KdV, ultra-discrete Toda, and discrete Toda equations. It proves that each admits a unique solution to the initial value problem precisely when the initial configuration belongs to a class constructed via path encodings, and this class contains the supports of many shift-ergodic measures. The same path-encoding device produces a unified dynamical description that generalizes Pitman's transformation of reflection in the past maximum, identifies a natural carrier process, and establishes all-time reversibility for all four systems. It also shows that bi-infinite ultra-discrete solutions arise directly as ultra-discretizations of the corresponding discrete solutions.

Core claim

For each of the four equations, the initial value problem possesses a unique solution on the bi-infinite lattice whenever the initial data belongs to the class defined by the path encoding; the evolution of every such solution is completely characterized by a generalization of Pitman's transformation, the same device simultaneously supplies a carrier process and confirms that the flow is reversible in both time directions, and the ultra-discrete solutions are obtained by ultra-discretization from the discrete ones.

What carries the argument

The path encoding, an antiderivative of the configuration that converts the local lattice map into a global transformation generalizing Pitman's reflection in the past maximum.

If this is right

  • Unique all-time solutions exist for every initial datum in the class for each of the four equations.
  • The dynamics of every solution in the class are given by the generalized Pitman transformation.
  • A natural carrier process is associated to every configuration in the class.
  • All four systems are reversible for all time on the given class.
  • Bi-infinite ultra-discrete solutions arise as ultra-discretizations of the corresponding discrete solutions.

Where Pith is reading between the lines

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

  • The same path-encoding method could be applied directly to other integrable lattice maps defined by local rules.
  • Shift-ergodic measures supported on the class furnish stationary processes whose long-time statistics are governed by the carrier process.
  • The explicit reversibility might be used to construct invariant measures for the infinite-lattice dynamics.
  • Numerical schemes based on the path encoding could be tested for stability on large but finite truncations of the bi-infinite line.

Load-bearing premise

The initial data must belong to the specific class whose definition makes the path-encoding construction produce uniqueness.

What would settle it

An explicit bi-infinite configuration inside the stated class for which two distinct solutions exist (or none exists) at some positive time.

read the original abstract

We define bi-infinite versions of four well-studied discrete integrable models, namely the ultra-discrete KdV equation, the discrete KdV equation, the ultra-discrete Toda equation, and the discrete Toda equation. For each equation, we show that there exists a unique solution to the initial value problem when the given data lies within a certain class, which includes the support of many shift ergodic measures. Our unified approach, which is also applicable to other integrable systems defined locally via lattice maps, involves the introduction of a path encoding (that is, a certain antiderivative) of the model configuration, for which we are able to describe the dynamics more generally than in previous work on finite size systems, periodic systems and semi-infinite systems. In particular, in each case we show that the behaviour of the system is characterized by a generalization of the classical 'Pitman's transformation' of reflection in the past maximum, which is well-known to probabilists. The picture presented here also provides a means to identify a natural 'carrier process' for configurations within the given class, and is convenient for checking that the systems we discuss are all-time reversible. Finally, we investigate links between the different systems, such as showing that bi-infinite all-time solutions for the ultra-discrete KdV (resp. Toda) equation may appear as ultra-discretizations of corresponding solutions for the discrete KdV (resp. Toda) equation.

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 paper introduces bi-infinite versions of the ultra-discrete KdV, discrete KdV, ultra-discrete Toda, and discrete Toda equations. For each, it establishes existence and uniqueness of solutions to the initial-value problem when initial data belong to a specified admissible class (containing supports of many shift-ergodic measures). The proofs rely on a path-encoding construction (an antiderivative of the configuration) together with a generalized Pitman reflection map; the same framework yields all-time reversibility and a natural carrier process. Links between the ultra-discrete and discrete systems via ultra-discretization are also examined.

Significance. If the central claims hold, the work supplies a unified, probabilistically motivated approach to bi-infinite discrete integrable systems that extends earlier finite, periodic and semi-infinite results. The explicit connection to Pitman’s transformation and the identification of carrier processes are likely to be of interest both to the integrable-systems community and to researchers working on interacting particle systems and ergodic theory on lattices.

major comments (2)
  1. [§3] §3 (definition of the admissible class C): the manuscript asserts that C contains the supports of many shift-ergodic measures, yet the summability conditions required for the bi-infinite path encoding to be globally well-defined (zero drift or controlled growth of partial sums in both directions) are not separately verified for typical ergodic measures. Without this check, it remains unclear whether the uniqueness statement applies to the full class advertised in the abstract.
  2. [Theorem 4.2] Theorem 4.2 (uniqueness via generalized Pitman map): the proof that the path-encoding map is bijective on the admissible class relies on the antiderivative being well-defined everywhere; if the summability condition fails for some ergodic support inside the claimed class, the map ceases to be invertible and the uniqueness claim is compromised.
minor comments (2)
  1. [§2] Notation for the four systems is introduced piecemeal; a single comparative table listing the local update rules, path encodings, and Pitman maps would improve readability.
  2. [Introduction] Several references to earlier finite/semi-infinite results are given only by author names; full citations should be supplied at first mention.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the careful reading and constructive comments on the admissible class and uniqueness result. We address each major comment below and will revise the manuscript to incorporate explicit verification.

read point-by-point responses

  1. Referee: [§3] §3 (definition of the admissible class C): the manuscript asserts that C contains the supports of many shift-ergodic measures, yet the summability conditions required for the bi-infinite path encoding to be globally well-defined (zero drift or controlled growth of partial sums in both directions) are not separately verified for typical ergodic measures. Without this check, it remains unclear whether the uniqueness statement applies to the full class advertised in the abstract.

    Authors: The admissible class C is defined in §3 exactly as the set of configurations satisfying the summability conditions (zero drift or controlled growth of partial sums in both directions) that make the bi-infinite path encoding globally well-defined. The claim that C contains supports of many shift-ergodic measures is based on the ergodic theorem: for shift-ergodic measures with zero mean (or suitable moments), the partial sums satisfy the required growth conditions almost surely. We acknowledge that an explicit verification for representative cases was omitted. In the revision we will add a short remark or subsection in §3 providing this check, e.g., for i.i.d. measures with mean zero, confirming that the uniqueness statement applies to the advertised class. revision: yes

  2. Referee: [Theorem 4.2] Theorem 4.2 (uniqueness via generalized Pitman map): the proof that the path-encoding map is bijective on the admissible class relies on the antiderivative being well-defined everywhere; if the summability condition fails for some ergodic support inside the claimed class, the map ceases to be invertible and the uniqueness claim is compromised.

    Authors: Theorem 4.2 establishes bijectivity of the generalized Pitman map strictly on the admissible class C, where the summability conditions hold by definition and the antiderivative is therefore well-defined everywhere. The map is constructed and inverted within this class. The potential failure of invertibility would occur only for configurations outside C. With the added explicit verification in §3 that typical ergodic supports lie inside C, the uniqueness claim remains intact for the class advertised in the abstract. revision: yes

Circularity Check

0 steps flagged

No significant circularity; new bi-infinite encodings and uniqueness derived independently

full rationale

The paper introduces bi-infinite path encodings as a new antiderivative construction for the configurations and derives the dynamics via a generalized Pitman reflection, showing uniqueness for data in a defined class. This class is specified to make the encoding and map well-defined, with properties proved directly from the construction rather than by fitting parameters or reducing to prior self-citations. The abstract and context indicate the central claims rest on explicit definitions and derivations for the bi-infinite case, extending but not presupposing the finite/semi-infinite results in a load-bearing circular manner. No quoted equations or steps reduce the uniqueness result to its inputs by construction.

Axiom & Free-Parameter Ledger

0 free parameters · 2 axioms · 1 invented entities

The central claims rest on the introduction of path encodings (treated as new objects) and standard background facts about discrete lattice maps and ergodic measures; no numerical fitting parameters appear.

axioms (2)
  • standard math Standard properties of antiderivatives on integer lattices and local lattice maps
    Path encoding is defined as a certain antiderivative of the model configuration.
  • domain assumption Existence of shift-ergodic measures whose supports lie in the admissible class
    The class is stated to include supports of many such measures.
invented entities (1)
  • Path encoding (bi-infinite version) no independent evidence
    purpose: To encode configurations so that dynamics reduce to a generalized Pitman transformation
    Introduced in the paper as the central new device for describing bi-infinite evolution.

pith-pipeline@v0.9.0 · 5805 in / 1484 out tokens · 33816 ms · 2026-05-24T14:37:57.282470+00:00 · methodology

discussion (0)

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

Reference graph

Works this paper leans on

58 extracted references · 58 canonical work pages · 1 internal anchor

  1. [1]

    Bertoin, An extension of Pitman’s theorem for spectrally positive L´ evy processes, Ann

    J. Bertoin, An extension of Pitman’s theorem for spectrally positive L´ evy processes, Ann. Probab. 20 (1992), no. 3, 1464–1483

    work page 1992

  2. [2]

    Biane, P

    P . Biane, P . Bougerol, and N. O’Connell, Littelmann paths and Brownian paths, Duke Math. J. 130 (2005), no. 1, 127–167

    work page 2005

  3. [3]

    , Continuous crystal and Duistermaat-Heckman measure for Co xeter groups, Adv. Math. 221 (2009), no. 5, 1522–1583

    work page 2009

  4. [4]

    J. L. Bona and R. Smith, The initial-value problem for the Korteweg-de Vries equati on, Philos. Trans. Roy. Soc. London Ser. A 278 (1975), no. 1287, 555–601

    work page 1975

  5. [5]

    Bourgain, Periodic Korteweg de Vries equation with measures as initia l data, Selecta Math

    J. Bourgain, Periodic Korteweg de Vries equation with measures as initia l data, Selecta Math. (N.S.) 3 (1997), no. 2, 115–159

    work page 1997

  6. [6]

    D. A. Croydon, T. Kato, M. Sasada, and S. Tsujimoto, Dynamics of the box-ball system with random initial con- ditions via Pitman’s transformation , to appear in Mem. Amer. Math. Soc., preprint appears at arXi v:1806.02147, 2018

    work page internal anchor Pith review Pith/arXiv arXiv 2018

  7. [7]

    D. A. Croydon and M. Sasada, Detailed balance and invariant measures for discrete KdV- a nd Toda-type sys- tems, preprint appears at arXiv:2007.06203, 2020

    work page arXiv 2007

  8. [8]

    , Duality between box-ball systems of finite box and/or carrie r capacity, RIMS Kˆ okyˆ uroku BessatsuB79 (2020), 63–107

    work page 2020

  9. [9]

    , Discrete integrable systems and Pitman’s transformation , Stochastic analysis, random fields and inte- grable probability—Fukuoka 2019, Adv. Stud. Pure Math., vol. 87, Math. Soc. Japan, Tokyo, 2021, pp. 381–402

    work page 2019

  10. [10]

    D. A. Croydon, M. Sasada, and S. Tsujimoto, Dynamics of the ultra-discrete Toda lattice via Pitman’s tr ansfor- mation, RIMS Kˆ okyˆ uroku BessatsuB78 (2020), 235–250. BI-INFINITE SOLUTIONS FOR KDV - AND TODA-TYPE DISCRETE INTE GRABLE SYSTEMS 59

    work page 2020

  11. [11]

    Draief, J

    M. Draief, J. Mairesse, and N. O’Connell, Queues, stores, and tableaux , J. Appl. Probab. 42 (2005), no. 4, 1145–1167

    work page 2005

  12. [12]

    P . A. Ferrari, C. Nguyen, L. T. Rolla, and M. Wang, Soliton decomposition of the box-ball system , Forum Math. Sigma 9 (2021), Paper No. e60, 37

    work page 2021

  13. [13]

    B. M. Hambly, J. B. Martin, and N. O’Connell, Pitman’s 2M − X theorem for skip-free random walks with Markovian increments, Electron. Comm. Probab. 6 (2001), 73–77

    work page 2001

  14. [14]

    J. M. Harrison and R. J. Williams, On the quasireversibility of a multiclass Brownian service station, Ann. Probab. 18 (1990), no. 3, 1249–1268

    work page 1990

  15. [15]

    Hirota, Nonlinear partial difference equations

    R. Hirota, Nonlinear partial difference equations. I. A difference an alogue of the Korteweg-de Vries equation , J. Phys. Soc. Japan 43 (1977), no. 4, 1424–1433

    work page 1977

  16. [16]

    , Nonlinear partial difference equations. II. Discrete-tim e Toda equation, J. Phys. Soc. Japan 43 (1977), no. 6, 2074–2078

    work page 1977

  17. [17]

    , New solutions to the ultradiscrete soliton equations , Stud. Appl. Math. 122 (2009), no. 4, 361–376

    work page 2009

  18. [18]

    Hirota, S

    R. Hirota, S. Tsujimoto, and T. Imai, Difference scheme of soliton equations , Future Directions of Nonlinear Dynamics in Physical and Biological Systems (P . L. Christia nsen, J. C. Eilbeck, and R. D. Parmentier, eds.), Plenum, 1993, pp. 7–15

    work page 1993

  19. [19]

    Inoue, A

    R. Inoue, A. Kuniba, and T. Takagi, Integrable structure of box-ball systems: crystal, Bethe a nsatz, ultradis- cretization and tropical geometry, J. Phys. A 45 (2012), no. 7, 073001, 64

    work page 2012

  20. [20]

    Jeulin, Un th´ eor` eme de J

    T. Jeulin, Un th´ eor` eme de J. W. Pitman, S´ eminaire de Probabilit´ es, XIII (Univ. Strasbourg, Strasbourg, 1977/78), Lecture Notes in Math., vol. 721, Springer, Berlin, 1979, Wi th an appendix by M. Y or, pp. 521–532

    work page 1977

  21. [21]

    Kanki, J

    M. Kanki, J. Mada, and T. Tokihiro, Conserved quantities and generalized solutions of the ultr adiscrete KdV equation, J. Phys. A 44 (2011), no. 14, 145202, 13

    work page 2011

  22. [22]

    C. E. Kenig, G. Ponce, and L. V ega, W ell-posedness of the initial value problem for the Kortewe g-de Vries equation, J. Amer. Math. Soc. 4 (1991), no. 2, 323–347

    work page 1991

  23. [23]

    Killip, J

    R. Killip, J. Murphy, and M. Visan, Invariance of white noise for KdV on the line , Invent. Math. 222 (2020), no. 1, 203–282

    work page 2020

  24. [24]

    Kishimoto, W ell-posedness of the Cauchy problem for the Korteweg-de Vries equation at the critical regular- ity, Differential Integral Equations 22 (2009), no

    N. Kishimoto, W ell-posedness of the Cauchy problem for the Korteweg-de Vries equation at the critical regular- ity, Differential Integral Equations 22 (2009), no. 5-6, 447–464

    work page 2009

  25. [25]

    D. J. Korteweg and G. de Vries, On the change of form of long waves advancing in a rectangular canal, and on a new type of long stationary waves , Philos. Mag. (5) 39 (1895), no. 240, 422–443

    work page

  26. [26]

    Kotani, Construction of KdV flow – a unified approach , preprint appears at arXiv:2107.05428, 2021

    S. Kotani, Construction of KdV flow – a unified approach , preprint appears at arXiv:2107.05428, 2021

    work page arXiv 2021

  27. [27]

    I. M. Krichever, Algebraic curves and nonlinear difference equations, Uspekhi Mat. Nauk 33 (1978), no. 4(202), 215–216

    work page 1978

  28. [28]

    Kuniba, H

    A. Kuniba, H. Lyu, and M. Okado, Randomized box-ball systems, limit shape of rigged configur ations and thermodynamic Bethe ansatz, Nuclear Phys. B 937 (2018), 240–271

    work page 2018

  29. [29]

    O. E. Lanford III, J. L. Lebowitz, and E. H. Lieb, Time evolution of infinite anharmonic systems , J. Statist. Phys. 16 (1977), no. 6, 453–461

    work page 1977

  30. [30]

    Levine, H

    L. Levine, H. Lyu, and J. Pike, Double jump phase transition in a soliton cellular automato n, Int. Math. Res. Not. IMRN (2022), no. 1, 665–727

    work page 2022

  31. [31]

    Matsumoto and M

    H. Matsumoto and M. Y or, Some changes of probabilities related to a geometric Browni an motion version of Pitman’s 2M − X theorem , Electron. Comm. Probab. 4 (1999), 15–23

    work page 1999

  32. [32]

    , A version of Pitman’s 2M − X theorem for geometric Brownian motions , C. R. Acad. Sci. Paris S´ er. I Math. 328 (1999), no. 11, 1067–1074

    work page 1999

  33. [33]

    , An analogue of Pitman’s 2M − X theorem for exponential Wiener functionals. I. A time-inv ersion ap- proach, Nagoya Math. J. 159 (2000), 125–166

    work page 2000

  34. [34]

    , An analogue of Pitman’s 2M − X theorem for exponential Wiener functionals. II. The role o f the gener- alized inverse Gaussian laws , Nagoya Math. J. 162 (2001), 65–86

    work page 2001

  35. [35]

    J. Moser, Finitely many mass points on the line under the influence of an exponential potential–an integrable system, Dynamical systems, theory and applications (Rencontres, Battelle Res. Inst., Seattle, Wash., 1974), Springer, Berlin, 1975, pp. 467–497. Lecture Notes in Phys. , V ol. 38

    work page 1974

  36. [36]

    Nagai, D

    A. Nagai, D. Takahashi, and T. Tokihiro, Soliton cellular automaton, Toda molecule equation and sor ting algo- rithm, Phys. Lett. A 255 (1999), 265–271

    work page 1999

  37. [37]

    Nagai, T

    A. Nagai, T. Tokihiro, and J. Satsuma, The Toda molecule equation and the ε-algorithm, Math. Comp. 67 (1998), no. 224, 1565–1575. BI-INFINITE SOLUTIONS FOR KDV - AND TODA-TYPE DISCRETE INTE GRABLE SYSTEMS 60

    work page 1998

  38. [38]

    , Ultra-discrete Toda molecule equation, Phys. Lett. A 244 (1998), 383–388

    work page 1998

  39. [39]

    O’Connell, Random matrices, non-colliding processes and queues , S´ eminaire de Probabilit´ es, XXXVI, Lec- ture Notes in Math., vol

    N. O’Connell, Random matrices, non-colliding processes and queues , S´ eminaire de Probabilit´ es, XXXVI, Lec- ture Notes in Math., vol. 1801, Springer, Berlin, 2003, pp. 1 65–182

    work page 2003

  40. [40]

    , Directed polymers and the quantum Toda lattice , Ann. Probab. 40 (2012), no. 2, 437–458

    work page 2012

  41. [41]

    O’Connell and M

    N. O’Connell and M. Y or, Brownian analogues of Burke’s theorem , Stochastic Process. Appl. 96 (2001), no. 2, 285–304

    work page 2001

  42. [42]

    , A representation for non-colliding random walks , Electron. Comm. Probab. 7 (2002), 1–12

    work page 2002

  43. [43]

    J. W. Pitman, One-dimensional Brownian motion and the three-dimensiona l Bessel process , Advances in Appl. Probability 7 (1975), no. 3, 511–526

    work page 1975

  44. [44]

    Quastel and B

    J. Quastel and B. V alk´ o,KdV preserves white noise, Comm. Math. Phys. 277 (2008), no. 3, 707–714

    work page 2008

  45. [45]

    L. C. G. Rogers, Characterizing all diffusions with the 2M − X property, Ann. Probab. 9 (1981), no. 4, 561–572

    work page 1981

  46. [46]

    L. C. G. Rogers and J. W. Pitman, Markov functions, Ann. Probab. 9 (1981), no. 4, 573–582

    work page 1981

  47. [47]

    Saisho and H

    Y . Saisho and H. Tanemura, Pitman type theorem for one-dimensional diffusion process es, Tokyo J. Math. 13 (1990), no. 2, 429–440

    work page 1990

  48. [48]

    Sogo, Toda molecule equation and quotient-difference method , J

    K. Sogo, Toda molecule equation and quotient-difference method , J. Phys. Soc. Japan 62 (1993), no. 4, 1081– 1084

    work page 1993

  49. [49]

    Spohn, Generalized Gibbs ensembles of the classical Toda chain , J

    H. Spohn, Generalized Gibbs ensembles of the classical Toda chain , J. Stat. Phys. 180 (2020), no. 1-6, 4–22

    work page 2020

  50. [50]

    Takahashi and J

    D. Takahashi and J. Matsukidaira, On discrete soliton equations related to cellular automata , Phys. Lett. A 209 (1995), 184–188

    work page 1995

  51. [51]

    , Box and ball system with a carrier and ultra-discrete modified KdV equation, RIMS Kokyuroku (Kyoto University) 1020 (1997), 1–14

    work page 1997

  52. [52]

    Takahashi and J

    D. Takahashi and J. Satsuma, A soliton cellular automaton , J. Phys. Soc. Japan 59 (1990), 3514–3519

    work page 1990

  53. [53]

    Toda, Vibration of a chain with nonlinear interaction , Journal of the Physical Society of Japan 22 (1967), no

    M. Toda, Vibration of a chain with nonlinear interaction , Journal of the Physical Society of Japan 22 (1967), no. 2, 431–436

    work page 1967

  54. [54]

    Tokihiro, Ultradiscrete systems (cellular automata), Discrete integrable systems, Lecture Notes in Phys., vol

    T. Tokihiro, Ultradiscrete systems (cellular automata), Discrete integrable systems, Lecture Notes in Phys., vol. 644, Springer, Berlin, 2004, pp. 383–424

    work page 2004

  55. [55]

    Tokihiro, The mathematics of box-ball systems , Asakura Shoten, 2010

    T. Tokihiro, The mathematics of box-ball systems , Asakura Shoten, 2010

    work page 2010

  56. [56]

    Tokihiro, D

    T. Tokihiro, D. Takahashi, J. Matsukidaira, and J. Satsu ma, From soliton equations to integrable cellular au- tomata through a limiting procedure, Phys. Lett. A 76 (1996), 3247–3250

    work page 1996

  57. [57]

    Tsujimoto and R

    S. Tsujimoto and R. Hirota, Ultradiscrete KdV equation, J. Phys. Soc. Japan 67 (1998), no. 6, 1809–1810

    work page 1998

  58. [58]

    Willox, Y

    R. Willox, Y . Nakata, J. Satsuma, A. Ramani, and B. Gramma ticos, Solving the ultradiscrete KdV equation , Journal of Physics A: Mathematical and Theoretical 43 (2010), no. 48, 482003

    work page 2010