Pith. sign in

REVIEW 2 major objections 6 minor 36 references

Randomly forced KdVB equations are ergodic: their statistical state eventually forgets the initial condition and is unique.

Reviewed by Pith at T0; open to challenge. T0 means a machine referee read the full paper against a public rubric. the ladder, T0–T4 →

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

Under controllability and noise-growth conditions, the KdVB equation with localized or multiplicative white noise has a unique invariant measure, with exponential mixing in the localized case.

T0 review reviewed 2026-08-05 challenge →

load-bearing objection A serious, mostly sound paper that extends ergodicity theory to KdVB under localized and multiplicative noise; the main new estimates are credible, but Theorem 2.1 has a domain gap in the (LS) verification and a few deferred steps that need clean-up. the 2 major comments →

arxiv 2509.01921 v1 pith:LOEKLF3D submitted 2025-09-02 math.DS math.PR

Ergodicity for the randomly forced Korteweg-de Vries-Burgers equation

Peng Gao This is my paper
classification math.DS math.PR MSC 60H1535R6037A25
keywords ergodicityKorteweg-de Vries-Burgers equationCarleman estimateFoias-Prodi estimatedegenerate noisecoupling methodasymptotic couplinginvariant measure
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved

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 asks what happens, over long times, to solutions of the Korteweg-de Vries-Burgers equation when it is randomly forced by noise that is far from uniformly distributed: either concentrated in a small region of space-time, or multiplicative, meaning noise scaled by the solution value. It claims that in both cases the randomness eventually settles into a unique probability distribution on solutions, independent of where the solution started. For the space-time localized noise the convergence to this stationary measure is exponential in the number of noise periods; for multiplicative white noise the paper proves uniqueness of the invariant measure, and under a stronger growth condition, convergence of the law from any initial state. If correct, this means the long-term statistical behavior of randomly forced KdVB waves is fully determined by the forcing alone. The proof relies on new quantitative estimates for the underlying deterministic equation, in particular a Carleman estimate and a Foias-Prodi estimate.

Core claim

In the author's own terms, the contribution is three theorems. Theorem 2.1: for the KdVB equation on the circle with a T-periodic deterministic forcing h, driven by an i.i.d. space-time localized noise satisfying structural condition (DN) and approximate controllability condition (AC), there exists a unique stationary measure mu and positive constants C, sigma such that every initial condition u0 satisfies ||P_k(u0,.) - mu||*_L <= C(1 + ||u0||^2)e^{-sigma k}. Theorem 2.2: for multiplicative white noise g(u)dW with Lipschitz coefficient satisfying (g1)-(g3) and linear-growth constant L3 < 1, provided the noise touches at least M >= N0 modes, the Markov semigroup has a unique ergodic invariant

What carries the argument

The engine of the localized-noise proof is a new global Carleman estimate for the linear complex KdVB equation (Theorem 3.1). It gives weighted L2 bounds on a solution over the whole torus in terms of the equation's right-hand side and of the solution on an arbitrarily small subdomain omega; from it the paper derives an observability inequality and a truncated observability inequality (Theorem 3.2) saying that low modes of the initial data can be recovered from finitely many projected observations. This observability is fed into an optimal-control problem whose solution produces a 'squeezing' map: if two solutions start close, one can add a small control supported on finitely many modes so t

Load-bearing premise

For the localized-noise theorem, everything rests on Condition (AC): the noise must be able to steer any state in a large ball arbitrarily close to a fixed state using finitely many allowed random functions, and the paper proves this only for small deterministic forcing, so if the forcing is not small this precondition could fail and the exponential ergodicity proof would have no basis.

What would settle it

A direct test of Condition (AC): pick a specific noise support K, a target u-bar, and a ball BH(R), then compute the reachable set {S_l(v, zeta_1,...,zeta_l) : zeta_j in K}. If for some R there is a gap of size epsilon > 0 separating the reachable set from u-bar, then (AC) fails and Theorem 2.1 cannot be invoked. A concrete instance would be a non-small deterministic forcing h for which the KdVB equation has two attracting periodic responses; trajectories near the two responses could not be steered close to one common point by controls near zero, so no unique stationary measure should be expec

Watch this falsifier. Get emailed when new claim-graph text bears on it.

If this is right

  • If Theorem 2.1 holds for a given h and noise, then for any two initial data the transition laws approach the same stationary distribution exponentially, so prediction of long-run statistics such as means, correlations, and probabilities of large-amplitude waves does not require knowing the initial condition.
  • For small T-periodic deterministic forcing h, Proposition 4.3 verifies Condition (AC), so exponential ergodicity applies to the physically relevant case of small periodic pumping plus localized random shaking.
  • Under (g1)-(g3) with L3 < 1 and enough active noise modes, there is a unique ergodic invariant measure, so time averages of observables converge almost surely to that measure's expectation.
  • With the stricter growth bound L3 < 1/sqrt(5), even a single initial distribution converges to the invariant measure, making the model asymptotically stable in law.
  • The Foias-Prodi estimate gives explicit moment controls on the difference between true and nudged trajectories, quantifying how quickly information in high Fourier modes is forgotten.

Where Pith is reading between the lines

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

  • The same Carleman machinery is likely to yield exact controllability and quantitative decay results for KdVB-type equations, as the author hints; if so, the mixing theorem would follow from controllability alone in a wider range of forcings.
  • Condition (AC) is probably not necessary: ergodicity might persist for larger h even if approximate controllability to a single point fails, but the current proof would need a different route, such as controllability to a set or partial controllability on low modes.
  • The thresholds L3 < 1 and L3 < 1/sqrt(5) are likely not sharp; numerical experiments with multiplicative noise could map the actual boundary for loss of uniqueness, for example with a diffusion coefficient of the form g(u) = alpha u + c near alpha = 1.
  • The space-time-localized noise result can be read as evidence that deterministic KdVB transport amplifies finite-dimensional randomness into full ergodicity, suggesting similar results for other third-order dispersive SPDEs with Burgers-type dissipation.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

2 major / 6 minor

Summary. The paper studies the long-time statistical behavior of the randomly forced Korteweg-de Vries-Burgers (KdVB) equation on the one-dimensional torus. In the first part, for a space-time localized bounded noise of the form (2.2) satisfying Condition (DN) and an approximate controllability assumption (AC), Theorem 2.1 claims exponential mixing in the dual-Lipschitz metric. The proof combines a new global Carleman estimate (Theorem 3.1), an observability inequality (Proposition 3.1), a truncated observability inequality (Theorem 3.2), a squeezing property (Proposition 4.2), and the abstract coupling criterion of Shirikyan (Proposition 4.1). In the second part, for multiplicative white noise satisfying (g1)-(g3), Theorem 2.2 claims uniqueness of the invariant measure when L3<1, and Theorem 2.3 adds convergence to the invariant measure when L3<1/sqrt(5). These results rely on a Foias-Prodi estimate in expectation (Theorem 5.1), moment estimates, estimates in probability, and the asymptotic coupling method of Glatt-Holtz-Mattingly-Richards and Kulik-Scheutzow.

Significance. If correct, the results are significant: they would provide the first exponential ergodicity result for KdVB with degenerate space-time localized noise and the first ergodicity results for KdVB with multiplicative white noise. The new Carleman estimate and the Foias-Prodi estimate are potentially useful beyond this paper. The manuscript is largely self-contained on the analytic side, and the abstract probabilistic criteria are applied from the literature. The hypotheses are explicit, and the thresholds L3<1 and L3<1/sqrt(5) are concrete and verifiable. The main chain of reasoning from the PDE estimates to the abstract criteria is clearly laid out, except for the load-bearing gap discussed below.

major comments (2)
  1. [§4.3, definition of Φ] The verification of Hypothesis (LS) is not valid as written. The Markov transition in Theorem 2.1 is u_k = S(u_{k-1}, h + η_k), where h is a fixed deterministic force. Proposition 4.2 proves the contraction (4.9) between S(û0, h_ref) and S(u0, h_ref + Υ(h_ref, û0)(u0 − û0)). To satisfy (LS) of Proposition 4.1, the reference force h_ref must be the total force on the first path, namely h + η_k (or h + ζ for a control ζ), not η_k alone. The paper instead defines Φ(u0,û0,η) := Υ(η,û0)(u0 − û0), dropping the deterministic term h. Thus the controlled transition used in (LS) is not the one required for the actual noise-driven system. This is a genuine mismatch and leaves Theorem 2.1 unproved.
  2. [§4.2 and §4.3, domain of Proposition 4.2] Even after replacing η by the total force h+η, Proposition 4.2 is not directly applicable. Proposition 4.2 is stated for h ∈ H^2(D_T), while in Theorem 2.1 the deterministic force is only assumed to be in H^1_loc(R+×T), and the noise support K is only shown to be compact in H^1_0(D_T). Hence h+η is in general only in H^1(D_T), not H^2(D_T). No extension of Proposition 4.2 to H^1(D_T) is proved. Since (LS) is a necessary hypothesis of the abstract criterion Proposition 4.1, this domain mismatch is load-bearing. It may be repairable by strengthening Proposition 4.2, but as written the proof of Theorem 2.1 is incomplete.
minor comments (6)
  1. [§4.2, Proposition 4.2] In the Lipschitz continuity statement, the displayed formula reads '∥Υ(h1, ˆu1) − Ψ(h2, ˆu2)∥'; the second term should be Υ(h2, ˆu2), not Ψ.
  2. [§6.1, Theorem 6.1] The statement says 'ξu0,v0 ∈ ˆC(Pu0, Pz0)'; the second marginal should be Pv0. Also, in the definition of D and D_n^ε, 'u(n)' should presumably be 'y(n)'.
  3. [§6.3, Step 2 of Theorem 2.3] Step 2 proves convergence of the coupling using P(∥u(n) − ˜v(n)∥ > ε), but Theorem 5.1 controls E∥u(n) − v(n)∥^2 for v solving the nudged equation (5.2), not for the stopped-equation solution ˜v from (6.2). If ˜v is a typo for v, please correct it; otherwise the argument needs clarification.
  4. [§3.1, condition (3.2)] The condition '|ψ′| > 0' cannot hold on the entire torus for a C∞ periodic function; it should be formulated as '|ψ′| > 0 on T\ω' (or an analogous condition). The constructed example indicates this is intended, but the text should be unambiguous.
  5. [§2.2, notation after (2.4)] The phrase 'η_l = ζ_l (1 ≤ l ≤ k)' is confusing; it should likely be '1 ≤ i ≤ l' or similar. Also, in Condition (DN), 'assumptations' is a typo.
  6. [References] References [14] and [26] appear to be the same paper (Glatt-Holtz, Martinez, Richards); please remove the duplicate or cite different versions appropriately.

Circularity Check

0 steps flagged

No significant circularity: the ergodicity conclusions are derived from newly proven estimates and external coupling criteria; the suspect (LS) verification is a correctness gap, not a circular reduction.

full rationale

The central results (Theorem 2.1-2.3) are not obtained by fitting or by defining the conclusion into the assumptions. Theorem 2.1 is conditional on (DN)+(AC); (AC) is an explicit controllability hypothesis, and Prop. 4.3 only verifies it in a small-forcing example via the author's earlier [12]. That is a self-citation but it is not load-bearing for the main theorem: if the example failed, Theorem 2.1 would still stand as a conditional statement under (AC). The coupling criteria are external benchmark results (Prop. 4.1 from Shirikyan [33]; Thms 6.1-6.2 from [23],[24]), not author-authored. The analytical substance - Carleman estimate (Thm 3.1), observability (Prop 3.1), truncated observability (Thm 3.2), squeezing control (Prop 4.2), Foias-Prodi estimates (Thm 5.1) - is proved in the paper from the KdVB equation and does not use the target ergodicity as input. The skeptical concern about Theorem 2.1's verification of (LS) is a substantive correctness issue (Prop. 4.2 appears to provide contraction relative to a deterministic reference h rather than for two noise-driven paths from the same eta, and eta in H^1_0(D_T) may be outside the H^2(D_T) domain of Upsilon), but this is a failure to instantiate an abstract hypothesis, not a circularity: no step defines the desired measure or mixing rate in terms of itself. Likewise Lemma 4.1 and Prop. 4.3 delegate proofs to [32] and [12]; these are omitted-support/citation issues, not reductions by construction. Thus no circular step is exhibited, and the paper is best scored as essentially non-circular.

Axiom & Free-Parameter Ledger

0 free parameters · 4 axioms · 0 invented entities

No parameters are fitted to data; constants are structural (eigenvalues, ball radii). The paper's contribution is a set of new inequalities (Carleman, truncated observability, Foias-Prodi) and their consequences, not a new physical entity. The load-bearing axioms are standard stochastic-analysis tools plus the standing controllability and noise-growth hypotheses.

axioms (4)
  • domain assumption Backward uniqueness holds for the linear KdVB equation (3.1) on the torus
    Invoked without proof in Theorem 3.2, Step 1 to propagate chi v-hat = 0 from an open time interval to later times; the truncated observability inequality (3.17) depends on it.
  • domain assumption The stochastic system (1.1) with multiplicative noise is well-posed, Feller, and admits an invariant measure under (g1)-(g2) with L3 < 1
    Section 2.3 states this holds by similar arguments as in [25] and does not reproduce the derivation; Theorems 2.2-2.3 assume this background.
  • standard math Girsanov theorem applies to the infinite-dimensional measure changes used in Propositions 6.1 and Theorem 2.3
    Used to show mutual absolute continuity of laws of (5.1) and (6.2); the Novikov condition is verified with the shift (6.1).
  • domain assumption Standing hypotheses (DN), (AC), (g1)-(g3) and the compact-support structure of the localised noise
    These hypotheses define the random forcing; the theorems are conditional on them. (AC) is only verified for small deterministic h in Proposition 4.3.

reviewed 2026-08-05 · how reviews work

0 comments
Cite this review

Pith. "Pith review of Ergodicity for the randomly forced Korteweg-de Vries-Burgers equation." pith.science (2026). https://pith.science/paper/LOEKLF3D

@misc{pith2026250901921,
  author       = {Pith},
  title        = {Pith review of: Ergodicity for the randomly forced Korteweg-de Vries-Burgers equation},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/LOEKLF3D}},
  note         = {Machine review of arXiv:2509.01921}
}
Share X Bluesky LinkedIn Reddit HN
read the original abstract

Our goal in this paper is to investigate ergodicity of the randomly forced Korteweg-de Vries-Burgers(KdVB) equation driven by non-additive white noise. Under reasonable conditions, we show that exponential ergodicity for KdVB equation driven by a space-time localised noise and ergodicity for KdVB equation driven by a multiplicative white noise. Our proof is based on some newly developed analytical properties for KdVB equation, such as Carleman estimate, truncated observability inequality, Foia\c{s}-Prodi estimate. Combining these analytical properties with coupling method and asymptotic coupling method, we can investigate the long time behavior of randomly forced KdVB equation.

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

36 extracted references · 35 canonical work pages

  1. [1]

    Travelling waves for a non-local Korteweg-de Vries-Burgers equation[J]

    Achleitner F, Cuesta C M, Hittmeir S. Travelling waves for a non-local Korteweg-de Vries-Burgers equation[J]. Journal of Differential Equations, 2014, 257(3): 720-758

    work page 2014

  2. [2]

    Analysis and computation of a discrete KdV-Burgers type equation with fast dispersion and slow diffusion[J]

    Artstein Z, Gear C W, Kevrekidis I G, et al. Analysis and computation of a discrete KdV-Burgers type equation with fast dispersion and slow diffusion[J]. SIAM Journal on Numerical Analysis, 2011, 49(5): 2124-2143

    work page 2011

  3. [3]

    Controllability and Ergodicity of three dimensional Primitive Equations Driven by a Finite-Dimensional Force[J]

    Boulvard P M, Gao P, Nersesyan V. Controllability and Ergodicity of three dimensional Primitive Equations Driven by a Finite-Dimensional Force[J]. Archive for Rational Mechanics and Analysis, 2023, 247(1): 2

    work page 2023

  4. [4]

    Bona J L, Sun S M, Zhang B Y. Non-homogeneous boundary value problems for the Korteweg-de Vries and the Korteweg-de Vries-Burgers equations in a quarter plane[C]//Annales de l’Institut Henri Poincare (C) Non Linear Analysis. Elsevier Masson, 2008, 25(6): 1145-1185

    work page 2008

  5. [5]

    Generalized couplings and ergodic rates for SPDEs and other Markov models[J]

    Butkovsky O, Kulik A, Scheutzow M. Generalized couplings and ergodic rates for SPDEs and other Markov models[J]. 2020

    work page 2020

  6. [6]

    Bang-Bang Property for Time Optimal Control of the Korteweg-de Vries-Burgers Equation[J]

    Chen M. Bang-Bang Property for Time Optimal Control of the Korteweg-de Vries-Burgers Equation[J]. Applied Mathematics & Optimization, 2017, 76(2): 399-414

    work page 2017

  7. [7]

    Global well-posedness and exponential decay rates for a KdV-Burgers equation with indefinite damping[C]//Annales de l’Institut Henri Poincare (C) Non Linear Analysis

    Cavalcanti M M, Cavalcanti V N D, Komornik V, et al. Global well-posedness and exponential decay rates for a KdV-Burgers equation with indefinite damping[C]//Annales de l’Institut Henri Poincare (C) Non Linear Analysis. Elsevier Masson, 2014, 31(5): 1079-1100

    work page 2014

  8. [8]

    Stochastic Equations in Infinite Dimensions, Cambridge University Press, 2014

    Da Prato G., Zabczyk J. Stochastic Equations in Infinite Dimensions, Cambridge University Press, 2014

    work page 2014

  9. [9]

    Well-posedness of the stochastic KdV-Burgers equation[J]

    Richards G. Well-posedness of the stochastic KdV-Burgers equation[J]. Stochastic Processes and their Appli- cations, 2014, 124(4): 1627-1647

    work page 2014

  10. [10]

    Complex-Valued Burgers and KdV-Burgers Equations[J]

    Khanal N, Wu J, Yuan J M, et al. Complex-Valued Burgers and KdV-Burgers Equations[J]. Journal of nonlinear science, 2010, 20(3): 341-360

    work page 2010

  11. [11]

    Wiss., Band 181, Springer-Verlag, NewYork-Heidelberg, translated fromthe French by P.Kenneth(1972)

    Lions, J.L., Magenes, E., Non-Homogeneous Boundary Value Problems and Applications[M], vol.I, Grundlehren Math. Wiss., Band 181, Springer-Verlag, NewYork-Heidelberg, translated fromthe French by P.Kenneth(1972)

    work page 1972

  12. [12]

    Recurrent solutions of the derivative Ginzburg-Landau equation with boundary forces[J]

    Gao P. Recurrent solutions of the derivative Ginzburg-Landau equation with boundary forces[J]. Applicable Analysis, 2018, 97(16): 2743-2761

    work page 2018

  13. [13]

    Stratonovich-Khasminskii averaging principle for multiscale random Korteweg-de Vries-Burgers equa- tion[J]

    Gao P. Stratonovich-Khasminskii averaging principle for multiscale random Korteweg-de Vries-Burgers equa- tion[J]. Nonlinearity, 2023, 36(11): 6124

    work page 2023

  14. [15]

    Grad H., Hu P.N., Unified shock profile in a plasma, Phys. Mech. 10 (1967) 2596-2601

    work page 1967

  15. [16]

    Controllability of Evolution Equations, Korea, 1996

    Fursikov A V and Imanuvilov O Y. Controllability of Evolution Equations, Korea, 1996. 33

    work page 1996

  16. [17]

    Carleman estimates for global uniqueness, stability and numerical methods for coefficient inverse problems[J]

    Klibanov M V. Carleman estimates for global uniqueness, stability and numerical methods for coefficient inverse problems[J]. Journal of Inverse and Ill-Posed Problems, 2013, 21(4): 477-560

    work page 2013

  17. [18]

    Invariant measures for nonlinear SPDE’s: uniqueness and stability[J]

    Maslowski B, Seidler J. Invariant measures for nonlinear SPDE’s: uniqueness and stability[J]. Archivum Mathematicum, 1998, 34(1): 153-172

    work page 1998

  18. [19]

    An introduction to 3D stochastic fluid dynamics[J]

    Albeverio S, Flandoli F, Sinai Y G, et al. An introduction to 3D stochastic fluid dynamics[J]. SPDE in Hydrodynamic: Recent Progress and Prospects: Lectures given at the CIME Summer School held in Cetraro, Italy August 29-September 3, 2005, 2008: 51-150

    work page 2005

  19. [20]

    Sur le comportement global des solutions non-stationnaires des ´ equations de Navier-Stokes en dimension 2[J]

    Foias C, Prodi G. Sur le comportement global des solutions non-stationnaires des ´ equations de Navier-Stokes en dimension 2[J]. Rendiconti del Seminario matematico della Universita di Padova, 1967, 39: 1-34

    work page 1967

  20. [21]

    Exponential mixing for the white-forced damped nonlinear wave equation[J]

    Martirosyan D. Exponential mixing for the white-forced damped nonlinear wave equation[J]. Evolution Equations and Control Theory, 2014, 3(4): 645-670

    work page 2014

  21. [22]

    Ergodicity results for the stochastic Navier-Stokes equations: an introduction[J]

    Debussche A. Ergodicity results for the stochastic Navier-Stokes equations: an introduction[J]. Topics in Mathematical Fluid Mechanics: Cetraro, Italy 2010, Editors: Hugo Beirao da Veiga, Franco Flandoli, 2013: 23-108

    work page 2010

  22. [23]

    On unique ergodicity in nonlinear stochastic partial differential equations[J]

    Glatt-Holtz N, Mattingly J C, Richards G. On unique ergodicity in nonlinear stochastic partial differential equations[J]. Journal of Statistical Physics, 2017, 166: 618-649

    work page 2017

  23. [24]

    Generalized couplings and convergence of transition probabilities[J]

    Kulik A, Scheutzow M. Generalized couplings and convergence of transition probabilities[J]. Probability Theory and Related Fields, 2018, 171(1-2): 333-376

    work page 2018

  24. [25]

    Uniqueness of the invariant measure and asymptotic stability for the 2D Navier-Stokes equations with multiplicative noise[J]

    Ferrario B, Zanella M. Uniqueness of the invariant measure and asymptotic stability for the 2D Navier-Stokes equations with multiplicative noise[J]. Discrete & Continuous Dynamical Systems, 2023: 1-35

    work page 2023

  25. [26]

    On the long-time statistical behavior of smooth solutions of the weakly damped, stochastically-driven KdV equation[J]

    Glatt-Holtz N, Martinez V R, Richards G H. On the long-time statistical behavior of smooth solutions of the weakly damped, stochastically-driven KdV equation[J]. arXiv:2103.12942, 2021

    Pith/arXiv arXiv 2021

  26. [27]

    Mathematics of two-dimensional turbulence[M]

    Kuksin S, Shirikyan A. Mathematics of two-dimensional turbulence[M]. Cambridge University Press, 2012

    work page 2012

  27. [28]

    Exponential mixing for a class of dissipative PDEs with bounded degenerate noise[J]

    Kuksin S, Nersesyan V, Shirikyan A. Exponential mixing for a class of dissipative PDEs with bounded degenerate noise[J]. Geometric and Functional Analysis, 2020, 30: 126-187

    work page 2020

  28. [29]

    Mixing via controllability for randomly forced nonlinear dissipative PDEs[J]

    Kuksin S, Nersesyan V, Shirikyan A. Mixing via controllability for randomly forced nonlinear dissipative PDEs[J]. Journal de l’ ´Ecole polytechnique-Math´ ematiques, 2020, 7: 871-896

    work page 2020

  29. [30]

    Randomly driven Korteweg-de Vries-Burgers equation[J]

    Or lowski A. Randomly driven Korteweg-de Vries-Burgers equation[J]. Physical Review E, 1994, 49(3): 2465

    work page 1994

  30. [31]

    Exponential mixing for randomly forced partial differential equations: method of cou- pling[M]//Instability in Models Connected with Fluid Flows II

    Shirikyan A. Exponential mixing for randomly forced partial differential equations: method of cou- pling[M]//Instability in Models Connected with Fluid Flows II. New York, NY: Springer New York, 2008: 155-188

    work page 2008

  31. [32]

    Control and mixing for 2D Navier-Stokes equations with space-time localised noise[J]

    Shirikyan A. Control and mixing for 2D Navier-Stokes equations with space-time localised noise[J]. Ann. Sci. ´Ec. Norm. Sup´ er.(4), 2015, 48(2): 253-280

    work page 2015

  32. [33]

    Controllability implies mixing II

    Shirikyan A. Controllability implies mixing II. Convergence in the dual-Lipschitz metric[J]. Journal of the European Mathematical Society, 2021, 23(4): 1381-1422

    work page 2021

  33. [34]

    Su C.H., Gardner C.S., Derivation of the Korteweg-de Vries and Burgers equation, J. Math. Phys. 10 (1969) 536-539

    work page 1969

  34. [35]

    Carleman estimates and unique continuation for solutions to boundary value problems J

    Tataru D. Carleman estimates and unique continuation for solutions to boundary value problems J. Math. Pures Appl. 75 (1996), 367-408

    work page 1996

  35. [36]

    Carleman estimates for parabolic equations and applications[J]

    Yamamoto M. Carleman estimates for parabolic equations and applications[J]. Inverse problems, 2009, 25(12): 123013

    work page 2009

  36. [37]

    Weakly damped KdV soliton dynamics with the random force[J]

    Zahibo N, Pelinovsky E, Sergeeva A. Weakly damped KdV soliton dynamics with the random force[J]. Chaos, Solitons & Fractals, 2009, 39(4): 1645-1650. 34

    work page 2009

This paper was first reviewed by deepseek-v4-flash on August 5, 2026.