pith. sign in

arxiv: 2606.24473 · v1 · pith:2IS3B2DUnew · submitted 2026-06-23 · 💻 cs.MS · cs.NA

New convergence results for Carleman linearization

Michele Boreale , Luisa Collodi This is my paper

Pith reviewed 2026-06-25 21:46 UTC · model grok-4.3

classification 💻 cs.MS cs.NA
keywords Carleman linearizationerror boundspolynomial ODEsDyson-Duhamel expansiongeometric convergencetruncation errorStuart-Landau systemVan der Pol oscillator
0
0 comments X

The pith

Finite Carleman truncations of polynomial ODEs admit explicit error bounds ensuring geometric convergence over certified time horizons.

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

The paper establishes new error bounds for finite truncations in the Carleman linearization of polynomial ordinary differential equations. It works directly with the monomial basis for observables like state coordinates and employs a Dyson-Duhamel expansion to isolate the linear dynamics from nonlinear interactions. This yields degree-aware bounds that preserve information from the original linear part and demonstrate geometric convergence within explicit time horizons. Comparisons on systems like Stuart-Landau and Van der Pol show improvements over previous bounds such as those by Forets and Pouly. A sympathetic reader would care because these bounds enable more reliable certified approximations for nonlinear systems without relying on abstract norms.

Core claim

We prove new error bounds for finite Carleman truncations of polynomial ordinary differential equations. The analysis works directly in the original monomial basis and for selected observables, such as state coordinates. Using a Dyson--Duhamel expansion, we separate the degree-preserving linear part from the degree-raising nonlinear part and track how truncation errors can propagate back to the observable. The resulting bounds are degree-aware and retain logarithmic-norm information from the original linear dynamics. We obtain explicit finite-degree estimates and geometric convergence over certified time horizons. Comparisons with existing bounds, in particular those of Forets--Pouly, are gi

What carries the argument

Dyson-Duhamel expansion separating the degree-preserving linear part from the degree-raising nonlinear part in the monomial basis

If this is right

  • Explicit finite-degree error estimates become available for polynomial ODE approximations.
  • Geometric convergence of truncation error holds over certified time horizons.
  • The bounds retain logarithmic-norm information from the linear dynamics.
  • Improved error performance appears on the Stuart-Landau and Van der Pol systems relative to prior bounds.

Where Pith is reading between the lines

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

  • The degree-aware bounds could support adaptive selection of truncation degree in numerical solvers.
  • Similar expansion techniques might apply to other linearization approaches for dynamical systems.
  • Certified horizons could inform verification methods in control or simulation software.
  • Extensions to higher-dimensional polynomial systems would test the scalability of the estimates.

Load-bearing premise

The ordinary differential equation must be polynomial, allowing separation of linear and nonlinear terms in the monomial basis via the Dyson-Duhamel expansion.

What would settle it

A numerical computation on a polynomial ODE where the observed truncation error exceeds the derived bound inside the certified time horizon would disprove the result.

Figures

Figures reproduced from arXiv: 2606.24473 by Luisa Collodi, Michele Boreale.

Figure 1
Figure 1. Figure 1: Van der Pol system: error ratios (bound in Theorem 2 [PITH_FULL_IMAGE:figures/full_fig_p012_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: Stuart-Landau system: error ratios (bound in Theo [PITH_FULL_IMAGE:figures/full_fig_p014_2.png] view at source ↗
read the original abstract

We prove new error bounds for finite Carleman truncations of polynomial ordinary differential equations. The analysis works directly in the original monomial basis and for selected observables, such as state coordinates. Using a Dyson--Duhamel expansion, we separate the degree-preserving linear part from the degree-raising nonlinear part and track how truncation errors can propagate back to the observable. The resulting bounds are degree-aware and retain logarithmic-norm information from the original linear dynamics. We obtain explicit finite-degree estimates and geometric convergence over certified time horizons. Comparisons with existing bounds, in particular those of Forets--Pouly, are given on the Stuart--Landau and Van der Pol systems.

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

0 major / 3 minor

Summary. The paper proves new error bounds for finite Carleman truncations of polynomial ordinary differential equations. The analysis works directly in the monomial basis for selected observables such as state coordinates. Using a Dyson-Duhamel expansion, it separates the degree-preserving linear part from the degree-raising nonlinear part and tracks truncation error propagation back to the observable. The resulting bounds are degree-aware, retain logarithmic-norm information from the original linear dynamics, and yield explicit finite-degree estimates with geometric convergence over certified time horizons. Comparisons with existing bounds, in particular those of Forets-Pouly, are provided on the Stuart-Landau and Van der Pol systems.

Significance. If the central claims hold, the work strengthens the theoretical foundation for Carleman linearization by delivering explicit, degree-aware error bounds that preserve logarithmic-norm data and guarantee geometric convergence on certified intervals. The direct use of the monomial basis and the clean separation via Dyson-Duhamel expansion are technically useful for verification and control applications. The explicit comparisons on standard benchmark systems add concrete evidence of improvement over prior bounds.

minor comments (3)
  1. The abstract and introduction should explicitly state the precise polynomial degree range for which the separation holds without additional assumptions on the linear part.
  2. In the comparison section, the time horizons and truncation degrees used for the Stuart-Landau and Van der Pol examples should be listed in a table for direct reproducibility of the reported error curves.
  3. A short remark on how the logarithmic norm is computed numerically for the linear part would help readers implement the bounds.

Simulated Author's Rebuttal

0 responses · 0 unresolved

We thank the referee for the positive summary, significance assessment, and recommendation of minor revision. The referee's description of the manuscript is accurate. No major comments were provided in the report.

Circularity Check

0 steps flagged

No significant circularity; derivation self-contained

full rationale

The paper derives explicit finite-degree error bounds and geometric convergence for Carleman truncations of polynomial ODEs directly from the Dyson-Duhamel expansion applied in the monomial basis. This separates the degree-preserving linear dynamics from degree-raising nonlinear terms and tracks truncation error propagation to observables such as state coordinates, retaining logarithmic-norm information. The resulting bounds are compared against prior work (e.g., Forets-Pouly) on concrete systems but do not reduce any claimed prediction or uniqueness result to a fit, self-citation chain, or definitional equivalence. No load-bearing step invokes an ansatz or theorem whose justification collapses to the present paper's inputs.

Axiom & Free-Parameter Ledger

0 free parameters · 2 axioms · 0 invented entities

The work rests on standard mathematical tools from ODE theory and linear algebra; no free parameters, new entities, or ad-hoc axioms are introduced in the abstract.

axioms (2)
  • standard math Properties of logarithmic norms for the linear dynamics
    Invoked to retain information from the original linear part in the error bounds.
  • domain assumption Validity of the Dyson-Duhamel expansion for separating degree-preserving and degree-raising terms
    Central to tracking how truncation errors propagate back to the observable.

pith-pipeline@v0.9.1-grok · 5627 in / 1287 out tokens · 28970 ms · 2026-06-25T21:46:54.371105+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

32 extracted references · 1 linked inside Pith

  1. [1]

    Reachability analysis of nonlinear sy stems using conservative polynomial- ization and non-convex sets

    Matthias Althoff. Reachability analysis of nonlinear sy stems using conservative polynomial- ization and non-convex sets. In Proceedings of the 16th International Conference on Hybrid Systems: Computation and Control , pages 173–182. ACM, 2013

    2013

  2. [2]

    An introduction to CORA 2015

    Matthias Althoff. An introduction to CORA 2015. In Proceedings of the Workshop on Applied Verification for Continuous and Hybrid Systems , pages 120–151, 2015

    2015

  3. [3]

    Set p ropagation techniques for reach- ability analysis

    Matthias Althoff, Goran Frehse, and Antoine Girard. Set p ropagation techniques for reach- ability analysis. Annual Review of Control, Robotics, and Autonomous Systems , 4:369–395, 2021

    2021

  4. [4]

    Error bounds for C arleman linearization of general nonlinear systems

    Arash Amini, Qiyu Sun, and Nader Motee. Error bounds for C arleman linearization of general nonlinear systems. In Proceedings of the Conference on Control and its Applicatio ns, pages 1–8. SIAM, 2021. 23

    2021

  5. [5]

    Carle man linearization of nonlinear systems and its finite-section approximations

    Arash Amini, Cong Zheng, Qiyu Sun, and Nader Motee. Carle man linearization of nonlinear systems and its finite-section approximations. Discrete and Continuous Dynamical Systems – Series B , 30(2):577–603, 2025

    2025

  6. [6]

    Reachability of Koopman linearized syst ems using random Fourier fea- ture observables and polynomial zonotope refinement

    Stanley Bak, Sergiy Bogomolov, Brandon Hencey, Niklas Koch dumper, Eric Lew, and Kos- tiantyn Potomkin. Reachability of Koopman linearized syst ems using random Fourier fea- ture observables and polynomial zonotope refinement. In Computer Aided Verification , Lecture Notes in Computer Science, pages 490–510. Springer , 2022

    2022

  7. [7]

    Richard Bellman and J. M. Richardson. On some questions ar ising in the approximate solution of nonlinear differential equations. Quarterly of Applied Mathematics , 20(4):333– 339, 1963

    1963

  8. [8]

    Berry, Andrew M

    Dominic W. Berry, Andrew M. Childs, Aaron Ostrander, and G uoming Wang. Quantum algorithm for linear differential equations with exponenti ally improved dependence on pre- cision. Communications in Mathematical Physics , 356:1057–1081, 2017

    2017

  9. [9]

    JuliaReach: A toolbox for set-based reachabili ty

    Sergiy Bogomolov, Marcelo Forets, Goran Frehse, Kostian tyn Potomkin, and Christian Schilling. JuliaReach: A toolbox for set-based reachabili ty. In Proceedings of the 22nd ACM International Conference on Hybrid Systems: Computati on and Control , pages 39–

  10. [10]

    Algorithms for exact and approximate l inear abstractions of polynomial continuous systems

    Michele Boreale. Algorithms for exact and approximate l inear abstractions of polynomial continuous systems. In Proceedings of the 21st International Conference on Hybrid Systems: Computation and Control , pages 1–10. ACM, 2018

    2018

  11. [11]

    Linearization, mode l reduction and reachability in nonlinear ODEs

    Michele Boreale and Luisa Collodi. Linearization, mode l reduction and reachability in nonlinear ODEs. In Reachability Problems 2022, volume 13608 of Lecture Notes in Computer Science, pages 49–66. Springer, 2022

    2022

  12. [12]

    Linearization, mode l reduction and reachability in nonlinear ODEs

    Michele Boreale and Luisa Collodi. Linearization, mode l reduction and reachability in nonlinear ODEs. Formal Methods in System Design , 66:419–454, 2025

    2025

  13. [13]

    Application de la théorie des équati ons intégrales linéaires aux systèmes d’équations différentielles non linéaires

    Torsten Carleman. Application de la théorie des équati ons intégrales linéaires aux systèmes d’équations différentielles non linéaires. Acta Mathematica, 59:63–87, 1932

    1932

  14. [14]

    Reachability Analysis of Non-linear Hybrid Systems Using T aylor Models

    Xin Chen. Reachability Analysis of Non-linear Hybrid Systems Using T aylor Models. PhD thesis, R WTH Aachen University, 2015

    2015

  15. [15]

    Taylor model flowpipe con- struction for non-linear hybrid systems

    Xin Chen, Erika Ábrahám, and Sriram Sankaranarayanan. Taylor model flowpipe con- struction for non-linear hybrid systems. In Proceedings of the IEEE Real-Time Systems Symposium, pages 183–192. IEEE, 2012

    2012

  16. [16]

    Flow*: An analyzer for non- linear hybrid systems

    Xin Chen, Erika Ábrahám, and Sriram Sankaranarayanan. Flow*: An analyzer for non- linear hybrid systems. In Computer Aided Verification , volume 8044 of Lecture Notes in Computer Science, pages 258–263. Springer, 2013

    2013

  17. [17]

    Ordinary Differential Equations with Applications , volume 34 of Texts in Applied Mathematics

    Carmen Chicone. Ordinary Differential Equations with Applications , volume 34 of Texts in Applied Mathematics. Springer, third edition, 2024

    2024

  18. [18]

    Pedro C. S. Costa, Philipp Schleich, Mauro E. S. Morales , and Dominic W. Berry. Further improving quantum algorithms for nonlinear differential eq uations via higher-order methods and rescaling. npj Quantum Information , 11:94, 2025

    2025

  19. [19]

    F. J. Dyson. The radiation theories of Tomonaga, Schwin ger, and Feynman. Physical Review, 75:486–502, 1949. 24

    1949

  20. [20]

    Explicit error bounds for Carleman linearization

    Marcelo Forets and Amaury Pouly. Explicit error bounds for Carleman linearization. arXiv preprint arXiv:1711.02552 , 2017

    Pith/arXiv arXiv 2017

  21. [21]

    Reachability of weakly nonlinear systems using Car- leman linearization

    Marcelo Forets and Christian Schilling. Reachability of weakly nonlinear systems using Car- leman linearization. In Reachability Problems, volume 13035 of Lecture Notes in Computer Science, pages 85–99. Springer, 2021

    2021

  22. [22]

    SpaceEx: Scalable verification of hybrid systems

    Goran Frehse, Colas Le Guernic, Alexandre Donzé, Scott Cotton, Rajarshi Ray, Olivier Lebeltel, Rodolfo Ripado, Antoine Girard, Thao Dang, and Od ed Maler. SpaceEx: Scalable verification of hybrid systems. In Computer Aided Verification , volume 6806 of Lecture Notes in Computer Science , pages 379–395. Springer, 2011

    2011

  23. [23]

    Reachability of uncertain linear syst ems using zonotopes

    Antoine Girard. Reachability of uncertain linear syst ems using zonotopes. In Hybrid Sys- tems: Computation and Control , volume 3414 of Lecture Notes in Computer Science , pages 291–305. Springer, 2005

    2005

  24. [24]

    Harrow, A vinatan Hassidim, and Seth Lloyd

    Aram W. Harrow, A vinatan Hassidim, and Seth Lloyd. Quan tum algorithm for linear systems of equations. Physical Review Letters , 103(15):150502, 2009

    2009

  25. [25]

    Nonlinear Dynamical Systems and Carleman Linearization

    Krzysztof Kowalski and Willi-Hans Steeb. Nonlinear Dynamical Systems and Carleman Linearization. World Scientific, 1991

    1991

  26. [26]

    Improved quantum algorithms for linear and nonlinear differential equations

    Hari Krovi. Improved quantum algorithms for linear and nonlinear differential equations. Quantum, 7:913, 2023

    2023

  27. [27]

    Krovi, Nuno F

    Jin-Peng Liu, Herman Øie Kolden, Hari K. Krovi, Nuno F. L oureiro, Konstantina Trivisa, and Andrew M. Childs. Efficient quantum algorithm for dissipa tive nonlinear differential equations. Proceedings of the National Academy of Sciences , 118(35):e2026805118, 2021

    2021

  28. [28]

    Rigorous integration of flo ws and ODEs using Taylor models

    Kyoko Makino and Martin Berz. Rigorous integration of flo ws and ODEs using Taylor models. In Proceedings of the 2009 Conference on Symbolic Numeric Comp utation, pages 79–84. ACM, 2009

    2009

  29. [29]

    Nedialkov, Kenneth R

    Nedialko S. Nedialkov, Kenneth R. Jackson, and George F . Corliss. Validated solutions of initial value problems for ordinary differential equations . Applied Mathematics and Com- putation, 105(1):21–68, 1999

    1999

  30. [30]

    Semigroups of Linear Operators and Applications to Partial Differential Equations, volume 44 of Applied Mathematical Sciences

    Amnon Pazy. Semigroups of Linear Operators and Applications to Partial Differential Equations, volume 44 of Applied Mathematical Sciences . Springer, 1983

    1983

  31. [31]

    Iterative Methods for Sparse Linear Systems

    Yousef Saad. Iterative Methods for Sparse Linear Systems . SIAM, second edition, 2003

    2003

  32. [32]

    Quantum a lgorithms for nonlinear dy- namics: Revisiting Carleman linearization with no dissipa tive conditions

    Hsuan-Cheng Wu, Jingyao Wang, and Xiantao Li. Quantum a lgorithms for nonlinear dy- namics: Revisiting Carleman linearization with no dissipa tive conditions. arXiv preprint arXiv:2405.12714, 2025. 25

    arXiv 2025