Pith. sign in

REVIEW 4 major objections 5 minor 33 references

Carleman-Fourier Linearization of Complex Dynamical Systems: Convergence and Explicit Error Bounds

T0 review · 4 major / 5 minor · reviewed 2026-08-12 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read The paper proves explicit exponential error bounds for a Fourier-exponential version of Carleman linearization of complex nonlinear systems with periodic vector fields, over a computable time horizon.

desk verdict Real extension of Carleman-Fourier linearization with explicit exponential bounds, but the advertised scope overstates what is proven for real periodic fields with Fourier decay radius R≤e. read the letter →

arxiv 2411.11598 v1 pith:FFFKNECY submitted 2024-11-18 math.DS cs.SYeess.SY

Panpan Chen , Nader Motee , Qiyu Sun This is my paper · ORCID
classification math.DScs.SYeess.SY MSC 37C1034C25
keywords Carleman-Fourierlinearizationperiodicvectorfieldsfinite-sectionapproximationexponentialconvergenceexpliciterrorboundscomplexdynamicalsystemsKuramotomodelmultiplefundamentalfrequencies
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

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

The reading

The paper introduces Carleman-Fourier linearization: instead of monomials, it lifts a complex nonlinear system $\dot{x}=g(t,x)$ with a periodic vector field into an infinite-dimensional linear system whose state variables are Fourier exponentials $e^{i\alpha x}$. For vector fields whose Fourier expansion is one-sided, meaning it contains no negative frequencies, the lifted matrix is block upper triangular, and the paper proves that the first block of the $N$-th finite-section approximation converges exponentially in $N$ to $e^{ix}$ over an explicit time horizon, with fully explicit constants. It also proves global exponential convergence when the constant Fourier coefficient of the vector field has strictly positive imaginary part. Because the error bounds are explicit, they can guide truncation-length choices in control, reachability, and quantum simulation. For vector fields that violate the one-sided condition, including the Kuramoto model, the paper introduces an augmented-state construction that restores analyticity at the cost of extra assumptions.

What carries the argument

The machinery is the Fourier-state lift: new variables $w_\alpha=e^{i\alpha x}$ for nonzero nonnegative multi-indices $\alpha$, grouped in blocks by total degree $k=1,\dots,N$. Under the analyticity condition $g_\alpha(t)=0$ for all negative multi-indices, the derivative of $w_\alpha$ only involves $w_\beta$ with $|\beta|\ge|\alpha|$, so the infinite matrix $B(t)$ is block upper triangular with diagonal blocks $i\operatorname{diag}(\alpha^T g_0(t))$. The finite-section system keeps the first $N$ blocks. The explicit error bound is obtained by writing the error $u_k=v_{k,N}-w_k$ in integral form with the diagonal kernel $K_k(t,s)=\exp(i\int_s^t \alpha^T g_0\,du)$, bounding the coupling blocks by the Schur-norm estimate $\|B_{k,l}\|_S\le D_0 k R^{k-l}$, and applying a discrete Gronwall-type inequality. The augmented state $[x^T,-x^T]^T$ is a second mechanism that manufactures the analyticity condition for general multi-frequency vector fields by pairing every negative index with a positive one.

What would settle it

For the system $\dot{x}=i(1-e^{ix})$ with initial $x_0=iy$, compute the $N$-truncated first block from (5.9) and compare $\max_{t\le T}|v_{1,N}(t)e^{-ix(t)}-1|$ with the right-hand side of (3.11); a discrepancy at fixed $N$, $T$ would refute the bound. Alternatively, adding a small term $\epsilon e^{-ix}$ to the vector field should break the block-triangular structure, so the observed error should stop shrinking exponentially in $N$, confirming that condition (1.9) is the load-bearing premise.

Watch

Extended reading notes

Core claim

The central discovery is that replacing monomials $x^{\alpha}$ by Fourier exponentials $e^{i\alpha x}$ in Carleman's lifting scheme converts a complex dynamical system $\dot{x}=g(t,x)$ with a one-sided periodic vector field into an infinite-dimensional linear system with a block upper-triangular matrix $B(t)$. For such systems, the first block $v_{1,N}$ of the $N$-th finite-section approximation satisfies the explicit bound $\max_j |v_{j,N}(t)e^{-ix_j(t)}-1| \le C_0 N^{-3/2} e^{D_0 t N} (e\|\exp(ix_0)\|_\infty/R)^{(e-1)N/(2e-1)}$ for $0\le t\le T^*_{CF}$; taking logarithms and writing $v_{j,N}=e^{i\xi_{j,N}}$ yields an approximation $\xi_{j,N}$ of the original state $x_j$ whose error is at most four times the same bound. When the constant Fourier coefficient $g_0$ has strictly positive imaginary part, the convergence extends to all $t\ge 0$ with the rate $((D_0+\mu_0)\|\exp(ix_0)\|_2/(\mu_0 R))^N$. For vector fields like Kuramoto's that have negative frequencies, the paper shows that lifting the augmented state $[x^T,-x^T]^T$ restores the hypothesis and proves the analogous exponential bound under the condition $R>e$ for real initial states.

Load-bearing premise

The central proofs require the periodic vector field to have only nonnegative Fourier frequencies, condition (1.9), which makes the lifted system triangular; when it fails, the paper's remedy only applies under the additional restriction that the Fourier decay radius exceed $e$ for real initial states.

Editorial extensions

If this is right

  • For systems satisfying condition (1.9) and Assumption 1.1, an order-$N$ truncation of the lifted system approximates $e^{ix(t)}$ with the explicit error bound of Theorem 3.1 on $[0,T^*_{CF}]$, so a desired tolerance directly fixes the truncation order.
  • Under the positive-imaginary condition (1.14) and the small-initial-state condition (1.15), the same finite-section approximation converges exponentially for all $t\ge 0$ with rate $((D_0+\mu_0)\|\exp(ix_0)\|_2/(\mu_0 R))^N$.
  • For vector fields with multiple fundamental frequencies that fail (1.9), the augmented state $[x^T,-x^T]^T$ restores the analyticity condition; with real initial states and $R>e$, exponential convergence holds on $[0,\widetilde{T}^*_{CF}]$ (Corollary 4.2).
  • When $1\le\|x_0\|_\infty<e^{-1}\ln R$, the Fourier method's guaranteed time horizon is at least as long and its convergence rate no worse than monomial Carleman linearization, as stated in (1.19).
  • The error bound depends on the initial state only through its imaginary parts, so the guaranteed accuracy is insensitive to how large the real parts of $x_0$ are.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • The formulas suggest a quantitative notion of a usable basin in complexified phase space: shifting a real initial state upward in the imaginary direction should extend the guaranteed horizon exactly as (1.11) predicts, a prediction one could test numerically on the example $\dot{x}=a(1-e^{ix})$.
  • The augmented-state doubling in Section 4 doubles the lifted dimension; for systems with additional symmetry, such as the zero-sum phases of the normalized Kuramoto model, a smaller symmetry-adapted basis might satisfy the analyticity condition with less overhead.
  • For quantum simulation of dissipative polynomial dynamics, the explicit $N$-dependence in Theorems 3.1 and 3.3 could be converted into a query or qubit count, though the paper only lists quantum computing as a motivation.
  • The comparison with real-valued systems suggests the real case has strictly better guaranteed horizons; extrapolating, a complex system whose vector field is real on the real axis should be approximated more efficiently by first separating real and imaginary parts rather than by the direct complex lift.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

4 major / 5 minor

Summary. The paper introduces Carleman-Fourier linearization for complex nonlinear dynamical systems with periodic vector fields. The method lifts the system using Fourier monomials e^{iα·x}; under the analyticity condition (1.9) the lifted matrix is block upper-triangular. Theorems 3.1 and 3.3 state explicit exponential error bounds for finite-section approximations on a finite horizon and on the entire time axis, respectively. Section 4 handles multiple fundamental frequencies by augmenting the state vector with its negative, giving Theorem 4.1 and Theorem 4.3. Section 5 presents numerical experiments for the scalar equation ẋ=a(1−be^{ix}) and for the Kuramoto model.

Significance. If the results are correct, the paper gives a useful extension of Carleman linearization to periodic vector fields, with explicit and computable error bounds, finite-section truncation criteria, and a natural treatment of multiple frequencies through state augmentation. The block upper-triangular observation and the global convergence result under positivity are interesting, and the numerical experiments support the qualitative claims. The main weaknesses are the heavy reliance on arguments and lemmas from the authors' earlier work [2], the omission of the proof of Theorem 2.1, a concrete incorrect constant bound in Section 4.2, and a scope limitation for real initial states when the Fourier decay radius R≤e that is not reflected in the abstract.

major comments (4)
  1. [Section 4.2, Eq. (4.16)] The asserted inequality C1 ≤ R^2/(2π e(e−1)) is false for the C1 defined in Theorem 4.1. Substituting the condition max|ℑ(ω_l x_{0,j})| < ln R − 1 into the displayed formula for C1 yields only C1 ≤ R^2/(√(2π) e(e−1)), and the stronger claimed bound fails, e.g., with R=4 and max|ℑ|=0.1 one obtains C1≈1.22 while R^2/(2π e(e−1))≈0.545. Since Corollary 4.2 and Eq. (4.19) use this simplification, the constants in those statements need to be corrected or the stronger inequality must be proved by a different argument.
  2. [Section 2, Theorem 2.1] Theorem 2.1 is stated with an explicit error bound but its proof is omitted: the text says the argument in [2] can be followed and the details are omitted. This theorem is listed as a contribution and is used in the comparison (1.19). The paper should either provide the proof in Section 6 or state precisely which theorem of [2] implies the complex-case bound with the same constants.
  3. [Section 6, Lemmas 6.2 and 6.3] Lemmas 6.2 and 6.3 are quoted from [2] without proof and are load-bearing for the proof of Theorem 3.1. Since the current paper advertises self-contained explicit error bounds, these lemmas should either be proved in the appendix or stated as cited theorems with exact references, so that a reader can verify the constants used in the main derivation.
  4. [Abstract and Section 4, Corollary 4.2(ii)] The abstract and introduction claim applicability to 'periodic vector fields' without qualification, but for real initial states the proved exponential convergence requires R>e (Corollary 4.2(ii)); for a real periodic vector field whose optimal Fourier decay radius satisfies R≤e, no exponential convergence of the finite-section approximation is established by the theorems in this paper. This is a genuine scope limitation, acknowledged only later in Section 4, and the claims in the abstract and in the comparison (1.19) should be qualified accordingly.
minor comments (5)
  1. [Theorem 3.1, Eq. (3.11)] The phrase 'exponential convergence' should be qualified: at t=T*_CF the factor e^{D0tN}(e∥exp(ix0)∥∞/R)^{(e−1)N/(2e−1)} equals 1, so the bound degenerates to O(N^{-3/2}) at the endpoint. Exponential-in-N convergence is established for t<T*_CF, as used in Corollary 3.2.
  2. [Eq. (3.11)] There is a typo in the displayed inequality: 'and and C0' contains a duplicated word.
  3. [Section 4.2 and Eq. (4.19)] The constant correction in the first major comment should be propagated to Corollary 4.2 and to the bound in Eq. (4.19), which currently inherit the incorrect simplification from Eq. (4.16).
  4. [Section 6.3] The proof of the comparison (1.19) is very sketchy and relies on numerical constants such as 4.9215 and 0.7076 without a complete derivation; a fully justified argument should be supplied.
  5. [References] Reference [26] is cited as 'In preparation' and is used for the comparison with real-system results; the paper should cite a published or otherwise publicly verifiable version, or the comparison should be proved directly.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the Carleman-Fourier error bounds are explicit functions of the stated assumptions and are not fitted or defined in terms of the predicted quantities.

full rationale

The paper's derivation chain is not circular. The central error bound (3.11) in Theorem 3.1 is an explicit function of the assumed Fourier-decay constants D0 and R, the initial condition x0, the truncation order N, and the time t; no parameter is fitted to the error that the bound is supposed to predict. The analyticity condition (1.9) is a structural assumption that forces the lifted matrix B(t) to be block upper triangular (3.5), and the proof independently controls w1(t)=e^{ix(t)} via Lemma 6.1 and then derives the finite-section error through Lemmas 6.2-6.3. Although Lemmas 6.2, 6.3 and the proof of Theorem 2.1 are cited from the authors' own [2], those are general ODE and combinatorial tools whose stated assumptions do not include the target result of this paper; under Rule 4 they count as independent support rather than circularity. Section 4's augmented-state construction removes (1.9), and the real-initial-state restriction R>e in Corollary 4.2(ii) is a genuine, explicitly acknowledged scope limitation, not a circular move: the bound is still proved from (4.2), (4.6), and (4.7) rather than assumed. The comparison with the unpublished [26] is contextual and not load-bearing for any theorem in this paper. There is no self-definition, no fitted input renamed as a prediction, and no uniqueness claim imported from the authors' prior work to force the choice of method. Thus the analysis is self-contained against the stated assumptions, and no circularity score above zero is warranted.

Assumptions & free parameters 0 free parameters · 5 assumptions · 0 invented entities

The central results rest on the three domain assumptions (Assumption 1.1, (1.9), (1.14)) and on technical lemmas quoted from the authors' earlier work [2]. No free parameters are fitted; the bounds are explicit in D0, R, mu0, and the initial state.

assumptions (5)
  • domain assumption Assumption 1.1: sup_t sum_{|alpha|=k} sum_j |g_{j,alpha}(t)| <= D0 R^{-k} for all k>=0
    Primary input condition for all theorems; guarantees convergence of the Fourier and Maclaurin expansions of the vector field.
  • domain assumption Analyticity condition (1.9): g_alpha(t)=0 for all alpha in Z^d \ Z^d_+ and t>=0
    Ensures the Carleman-Fourier state matrix B(t) is block upper-triangular (3.5a), which is the structural basis for the finite-section error analysis.
  • domain assumption Positivity condition (1.14): min_{1<=j<=d} Im g_{j,0}(t) >= mu0 > 0 for all t>=0
    Required for Theorem 3.3's global-in-time exponential convergence; forces imaginary parts of x(t) to diverge to +infinity.
  • standard math Lemmas 6.2 and 6.3 (taken from [2, Lemmas 5.3 and 5.4]) giving integral representations and combinatorial bounds for the finite-section error system
    Quoted without proof; they are the core technical estimates used in the proofs of Theorems 3.1 and 3.3.
  • standard math The standard inequality |z mod 2pi| <= 4epsilon for all z in C with |e^{iz}-1| <= epsilon <= 1/2
    Used in Corollaries 3.2 and 3.4 to convert errors in e^{ix} into errors in the phase x.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Carleman-Fourier Linearization of Complex Dynamical Systems: Convergence and Explicit Error Bounds." pith.science (2026). https://pith.science/paper/FFFKNECY

@misc{pith2026241111598,
  author       = {Pith},
  title        = {Pith review of: Carleman-Fourier Linearization of Complex Dynamical Systems: Convergence and Explicit Error Bounds},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/FFFKNECY}},
  note         = {Machine review of arXiv:2411.11598}
}
read the original abstract

This paper presents a Carleman-Fourier linearization method for nonlinear dynamical systems with periodic vector fields involving multiple fundamental frequencies. By employing Fourier basis functions, the nonlinear dynamical system is transformed into a linear model on an infinite-dimensional space. The proposed approach yields accurate approximations over extended regions around equilibria and for longer time horizons, compared to traditional Carleman linearization with monomials. Additionally, we develop a finite-section approximation for the resulting infinite-dimensional system and provide explicit error bounds that demonstrate exponential convergence to the original system's solution as the truncation length increases. For specific classes of dynamical systems, exponential convergence is achieved across the entire time horizon. The practical significance of these results lies in guiding the selection of suitable truncation lengths for applications such as model predictive control, safety verification through reachability analysis, and efficient quantum computing algorithms. The theoretical findings are validated through illustrative simulations.

Figures

Figures reproduced from arXiv: 2411.11598 by the authors.

Figure 1
Figure 1. Plotted are the vector fields a(1 − e ix) of the complex dynamical system (1.3) with a = 1 (left), a = i (middle) and a = −i (right), where −π ≤ ℜx ≤ π and −π/2 ≤ ℑx ≤ π/2. Trajectories on the left figure have parameters a = 1 and initial x0 = iln(1 − e aiπ/2 ) ≈ 0.7854 + 0.3466i (in black), −1/2 (in cyan) and −3/2 (in red). Presented in the middle are trajectories with a = i and x0 = iln(1−e aiπ/2 ) ≈ −0.2330i (in … view at source ↗
Figure 2
Figure 2. Plotted on the left is the function min{h(φ, t), 10}, −π/2 ≤ φ ≤ π/2, 0 ≤ t ≤ 5, where h is given in (5.12). Presented in the middle is the actual time range min(T ∗ (φ), 3), −π/2 ≤ φ ≤ 0, in (5.14), where ℑx0 = 0 (in green) and ℑx0 = 2 (in blue), and the time range T ∗ CF in Theorem 3.1 when ℑx0 = 0 (in red) and when ℑx0 = 2 (in magenta). Shown on the right is the requirement on the initial x0 for the exponential c… view at source ↗
Figure 3
Figure 3. Plotted on the top are the finite-section approximation errors max(min(ECF (x0, T∗ , N), 2), −5) of the Carleman-Fourier linearization, defined in (5.19), where −π/2 ≤ ϕ ≤ π/2 as the x-axis and −2 ≤ ℑx0 ≤ 2 as the y-axis, and level curve ECF (x0, T∗ , N) = 0 (in black) for N = 10 and T ∗ = 2 (left), 1/2 (middle) and 1/4 (right) respectively. Shown in the middle are max(min(ECF (x0, T∗ , N), 2), −5) with −2 ≤ ℜx0 ≤ 2… view at source ↗
Figures from the paper (2 more)
Figure 4
Figure 4. Figure 4: Plotted on the top row are the vector fields of the dynamical system (5.22) for −4π/3 ≤ θ1(0), θ2(0) ≤ 4π/3 and the shadowed regions on which the vector field has relatively small magnitude, where K˜ = −1 and (ω1, ω2) = (0, 0) (top left), (0, 1) (top middle), (0.5, 0.5…
Figure 5
Figure 5. Figure 5: Plotted are the approximation error EC(θ1(0), θ2(0), ω1, ω2, N, T), −4π/3 ≤ θ1(0), θ2(0) ≤ 4π/3, in (5.28) of the finite-section method to the Carleman linearization of the dynamical system (5.27), where K˜ = −1, N = 10, T = 0.5 and (ω1, ω2) = (0, 0)(left), (0, 1)(midd…

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

33 extracted references · 31 canonical work pages

  1. [2]

    Amini, C

    A. Amini, C. Zheng, Q. Sun and N. Motee, Carleman linearization of nonlinear systems and its finite-section approximations, Discrete and Continuous Dynamical Systems Series B , 30(2), 2025, pp. 577–603

    work page 2025

  2. [26]

    Motee and Q

    N. Motee and Q. Sun, On exponential convergence of Carleman-Fourier linearization of nonlinear real dynamical systems, In prepartion

    work page

  3. [1]

    Abudia, J

    M. Abudia, J. A. Rosenfeld and R. Kamalapurkar, Carleman lifting for nonlinear system identification with guaranteed error bounds, 2023 American Control Conference (ACC), San Diego, CA, USA , 2023, pp. 929–934

    work page 2023

  4. [3]

    Amini, Q

    A. Amini, Q. Sun and N. Motee, Error bounds for Carleman linearization of general nonlinear systems, 2021 Proc. the Conference on Control and its Applications , SIAM, 2021, pp. 1–8. 42 PANPAN CHEN, NADER MOTEE, AND QIYU SUN

    work page 2021

  5. [4]

    Amini, Q

    A. Amini, Q. Sun and N. Motee, Approximate optimal control design for a class of nonlinear systems by lifting Hamilton-Jacobi-Bellman equation, 2020 American Control Conference (ACC), IEEE, 2020, pp. 2717–2722

    work page 2020

  6. [5]

    Amini, Q

    A. Amini, Q. Sun and N. Motee, Quadratization of Hamilton-Jacobi-Bellman equation for near-optimal control of nonlinear systems, 59th IEEE Conference on Decision and Control (CDC), IEEE, 2020, pp. 731–736

    work page 2020

  7. [6]

    Akiba, Y

    T. Akiba, Y. Morii and K. Maruta, Carleman linearization approach for chemical kinetics integration toward quantum computation, Scientific Reports, 13(1), 2023, pp. 3935

    work page 2023

  8. [7]

    Brockett, The early days of geometric nonlinear control, Automatica, 50(9), 2014, pp

    R. Brockett, The early days of geometric nonlinear control, Automatica, 50(9), 2014, pp. 2203–2224

    work page 2014

Show all 33 references
  1. [8]

    J. C. Bronski, T. E. Carty and L. DeVille, Synchronisation conditions in the Kuramoto model and their relationship to seminorms, Nonlinearity, 34(8), 2021, pp. 5399–5433

    2021

  2. [9]

    S. L. Brunton, J. L. Proctor and J. N. Kutz, Discovering governing equations from data by sparse identification of nonlinear dynamical systems, Proceedings of the National Academy of Sciences, 113(15), 2016, pp. 3932–3937

    2016

  3. [10]

    Dietert, Stability and bifurcation for the Kuramoto model, Journal de Math´ ematiques Pures et Appliqu´ ees, 105(4), 2016, pp

    H. Dietert, Stability and bifurcation for the Kuramoto model, Journal de Math´ ematiques Pures et Appliqu´ ees, 105(4), 2016, pp. 451–489

    2016

  4. [11]

    Forets and A

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

    2017 arXiv

  5. [12]

    Forets and C

    M. Forets and C. Schilling, Reachability of weakly nonlinear systems using Carleman linearization, International Conference on Reachability Problems , Springer, 2021, pp. 85–99

    2021

  6. [13]

    Y. Guo, D. Zhang, Z. Li, Q. Wang and D. Yu, Overviews on the applications of the Kuramoto model in modern power system analysis, International Journal of Electrical Power & Energy Systems, 129, 2021, article no. 106804

    2021

  7. [14]

    Hashemian and A

    N. Hashemian and A. Armaou, Fast moving horizon estimation of nonlinear processes via Carleman linearization, 2015 American Control Conference (ACC) , IEEE, 2015, pp. 3379–3385

    2015

  8. [15]

    O. A. Heggli, J. Cabral, I. Konvalinka, P. Vuust and M. L. Kringelbach, A Kuramoto model of self-other integration across interpersonal synchronization strategies, PLoS Computational Biology, 15(10), 2019, article no. e1007422, 17pp

    2019

  9. [16]

    P. Ji, T. K. Peron, F. A. Rodrigues and J. Kurths, Low-dimensional behavior of Kuramoto model with inertia in complex networks, Scientific Reports, 4(1), 2014, article no. 4783

    2014

  10. [17]

    Korda and I

    M. Korda and I. Mezi´ c, Linear predictors for nonlinear dynamical systems: Koopman operator meets model predictive control, Automatica, 93, 2018, pp. 149–160

    2018

  11. [18]

    Korda and I

    M. Korda and I. Mezi´ c, On convergence of extended dynamic mode decomposition to the Koopman operator, Journal of Nonlinear Science , 28, 2018, pp. 687–710

    2018

  12. [19]

    Kowalski and W.H

    K. Kowalski and W.H. Steeb, Nonlinear dynamical systems and Carleman linearization, World Scientific, 1991

    1991

  13. [20]

    A. J. Krener, Linearization and bilinearization of control systems, Proceedings of the 1974 Allerton Conference on Circuit and Systems Theory, Urbana III , 1974

    1974

  14. [21]

    A. J. Krener, Bilinear and nonlinear realizations of input-output maps, SIAM Journal on Control, 13(4), 1975, pp. 827–834

    1975

  15. [22]

    Kuramoto, Chemical oscillations, waves, and turbulence, New York, Springer-Verlag, 1984

    Y. Kuramoto, Chemical oscillations, waves, and turbulence, New York, Springer-Verlag, 1984

    1984

  16. [23]

    J.P. Liu, H. O. Kolden, H. K. Krovi, N. F. Loureiro, K. Trivisa and A. W. Childs, Efficient quantum algorithm for dissipative nonlinear differential equations, Proceedings of the National Academy of Sciences, 118(35), 2021, article no. e2026805118

    2021

  17. [24]

    Loparo and G

    K. Loparo and G. Blankenship, Estimating the domain of attraction of nonlinear feedback systems, IEEE Transactions on Automatic Control , 23(4), 1978, pp. 602–608. CARLEMAN-FOURIER LINEARIZATION 43

    1978

  18. [25]

    Minisini, A

    J. Minisini, A. Rauh and E. P. Hofer, Carleman linearization for approximate solutions of nonlinear control problems: Part 1–theory, Proc. of the 14th Intl. Workshop on Dynamics and Control, 2007, pp. 215–222

    2007

  19. [27]

    Pruekprasert, J

    S. Pruekprasert, J. Dubut, T. Takisaka, C. Eberhart and A. Cetinkaya, Moment propagation through Carleman linearization with application to probabilistic safety analysis, 2022, arXiv preprint arXiv:2201.08648

    2022 arXiv

  20. [28]

    A. Rauh, J. Minisini and H. Aschemann, Carleman linearization for control and for state and disturbance estimation of nonlinear dynamical processes, IF AC Proceedings Volumes, 42(13), 2009, pp. 455–460

    2009

  21. [29]

    Rotondo, G

    D. Rotondo, G. Luta and J. H. U. Aarvag, Towards a Taylor-Carleman bilinearization approach for the design of nonlinear state-feedback controllers, European Journal of Control 68 , 2022, article no. 100670

    2022

  22. [30]

    W. H. Steeb and F. Wilhelm, Non-linear autonomous systems of differential equations and Carleman linearization procedure, Journal of Mathematical Analysis and Applications , 77(2), 1980, pp. 601–611

    1980

  23. [31]

    Surana, A

    A. Surana, A. Gnanasekaran and T. Sahai, An efficient quantum algorithm for simulating polynomial dynamical systems, Quantum Information Processing, 23(3), 2024, article no. 105, 22pp

    2024

  24. [32]

    Z. Wang, R. M. Jungers and C. J. Ong, Computation of invariant sets via immersion for discrete-time nonlinear systems, Automatica, 147, 2023, article no. 110686, 9pp

    2023

  25. [33]

    H. C. Wu, J. Wang and X. Li, Quantum algorithms for nonlinear dynamics: revisiting Carleman linearization with no dissipative conditions, 2024, arXiv preprint arXiv:2405.12714 . Department of Mathematics, University of Central Florida, Orlando, Florida 32816 Email address : pa...

    2024 arXiv

Pith tools

Reviewed August 12, 2026 · model on record in the stance chip above.