Pith. sign in

REVIEW 4 major objections 5 minor 22 references

Invariant measures for stochastic Burgers equation on unbounded domains

T0 review · 4 major / 5 minor · reviewed 2026-08-07 · deepseek-v4-flash

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

Pith's one-line read The paper proves that the damped stochastic Burgers equation with multiplicative colored noise on the real line has at least one invariant measure in $L^2(\mathbb{R})$ when $k>0$ and $a l^2 < \frac{3}{7}k$.

desk verdict Plausible extension of invariant-measure results for Burgers on R to multiplicative non-gradient noise, but the proof has two fixable gaps. read the letter →

arxiv 2506.07119 v1 pith:VSI2XJZW submitted 2025-06-08 math.DS math.PR

Zhenxin Liu , Zhiyuan Shi This is my paper
classification math.DSmath.PR MSC 35Q3535R6060H15
keywords stochasticBurgersequationunboundeddomaininvariantmeasuremultiplicativenoisemildsolutionuniformtailestimatestightnessKrylov-Bogolioubovtheorem
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 studies the long-time behavior of the damped stochastic Burgers equation on the whole real line, driven by multiplicative noise that is white in time and colored in space. It proves that a solution exists and is unique, that its expected norms stay bounded uniformly in time, and that the family of time-averaged distributions is tight. From tightness and the Feller property (continuity of the solution law in the initial data), the Krylov-Bogolioubov theorem produces at least one invariant measure on $L^2(\mathbb{R})$ under the smallness condition $a l^2 < \frac{3}{7}k$, where $a$ is the trace of the noise covariance and $l$ is the linear growth rate of the noise coefficient. This matters because an invariant measure is the mathematical notion of a stationary regime, so the result says that long-time statistics of the equation are well-defined on the unbounded spatial domain.

What carries the argument

The load-bearing object is the heat-kernel representation of the solution as a mild process, whose convolution terms $J_1$ and $J_2$ obey the Young-type bounds of Lemmas 3.1 and 3.2; these bounds make the truncated fixed-point map contractive and give local existence. The estimates that carry the argument are the Itô-formula energy inequality of Lemmas 3.5 and 4.1, which gives uniform-in-time $L^p$ and gradient bounds, and the cut-off tail estimate of Lemma 4.2, which uses a smooth function $\theta_m(x)=\theta(x/m)$ to show that the mass of the solution outside $\{|x|\ge m\}$ is small uniformly in time. Combining these with the compactness of Sobolev embeddings on bounded intervals produces a compact set on which the time-averaged distributions concentrate, i.e. tightness. This is the uniform tail-estimates method. The Feller property of the transition semigroup, proved by a stopping-time comparison of solutions starting from two nearby initial data, then activates the Krylov-Bogolioubov theorem, which converts tightness plus the Feller property into an invariant measure.

What would settle it

Numerically integrate (2.1) with $\sigma(u)=l u$, choose $k>0$ and $a$ with $a l^2<\frac{3}{7}k$, and record the empirical distributions of $u(t)$ over a long time interval; if these distributions do not converge to a fixed law, or if the solution's mass escapes to infinity, the theorem would be contradicted.

Watch

Extended reading notes

Core claim

The central claim is Theorem 2.6: for the equation $du=(u_{xx}-ku-\frac{1}{2}(u^2)_x)dt+\sigma(u)dW$ on $\mathbb{R}$ with $u_0\in L^2(\mathbb{R})$, if the noise coefficient $\sigma$ is Lipschitz with linear growth and the noise covariance satisfies $a l^2 < \frac{3}{7}k$, then the law of the solution has at least one invariant probability measure on $L^2(\mathbb{R})$. The proof shows first that a unique mild solution exists in $L^p(\mathbb{R})$ for $p\ge2$, then derives uniform-in-time energy bounds, and then uses a cut-off function to control the solution's mass outside large intervals uniformly in time. These uniform tail estimates make the family of time-averaged distributions tight, and the Feller property allows the Krylov-Bogolioubov theorem to produce a stationary measure. This extends earlier invariant-measure results for the stochastic Burgers equation on unbounded domains from additive noise to multiplicative noise.

Load-bearing premise

The argument assumes that the solution is smooth enough in space for Itô's formula to be applied directly; the paper says these calculations are formal and could be justified by an approximation, but it does not write out that approximation, so the energy and tail bounds rest on an unproved regularity assumption.

Editorial extensions

If this is right

  • If Theorem 2.6 is correct, the damped stochastic Burgers equation on $\mathbb{R}$ has at least one stationary probability distribution, so long-time statistical averages of the solution are well-defined.
  • The condition $a l^2 < \frac{3}{7}k$ gives a quantitative guide: the damping $k>0$ must dominate the combined intensity $a l^2$ of the multiplicative noise.
  • The uniform tail estimates show that, uniformly in time, the solution's mass concentrates on a compact region of space, which replaces the failing compactness of Sobolev embeddings on unbounded domains.
  • Because the theorem covers multiplicative noise that is white in time and colored in space, it broadens the class of stochastic Burgers models on the line for which a stationary regime is known to exist.
  • The result establishes existence only; uniqueness of the invariant measure remains open.

Reading between the lines

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

  • A sharper balance condition is likely possible: the factor $\frac{3}{7}$ comes from crude estimates in the tail lemma, and a refined argument might extend existence to larger noise-to-damping ratios.
  • The same tightness strategy should apply to related damped semilinear stochastic equations on $\mathbb{R}$ whenever an energy dissipation inequality like (4.2) holds, for example with different polynomial nonlinearities or other colored noises.
  • The threshold condition suggests a direction for numerical testing: simulations near $a l^2 = \frac{3}{7}k$ could indicate whether existence of invariant measures persists beyond the proved range.
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 studies the damped stochastic Burgers equation with multiplicative colored noise on the whole real line. It claims two main results: Theorem 2.3, existence and uniqueness of a mild solution in Lp(R) for p ≥ 2 with a uniform bound on the p-th moment, and Theorem 2.6, existence of an invariant measure on L2(R) under the assumptions k > 0 and a l^2 < 3/7 k. The proof strategy is standard: a truncated equation and fixed-point argument give local existence, stopping-time arguments and formal Itô-based energy estimates give global existence and uniform-in-time bounds, tail estimates provide tightness, and the Krylov-Bogolioubov theorem is applied after establishing the Feller property.

Significance. If the main theorem is correct, the paper gives a useful criterion for the existence of stationary distributions for the damped stochastic Burgers equation on an unbounded domain, going beyond gradient-form noise. The method is conventional and the paper contains no fitted parameters; the explicit condition a l^2 < 3/7 k is concrete and checkable. However, the current manuscript does not fully support the claims: two load-bearing arguments (the justification of the Itô-based energy estimates and the Feller property) contain gaps, and one displayed Itô formula is not correct as written. The overall strategy is sound and the gaps appear repairable, so the result is promising but not yet established.

major comments (4)
  1. [§3.2, Lemma 3.5; §4.1, Lemma 4.1] The energy estimates that drive global existence and the uniform-in-time bounds are obtained by applying Itô's formula to the mild solution as if it were a semimartingale in a Sobolev space. Lemma 3.5 states that the calculations are formal and "can be justified by a limiting procedure," but no such procedure is supplied, and Lemma 4.1 applies the same calculus without even that caveat. Since the solution is constructed only as a mild solution in Lp spaces, the hypotheses of the cited Itô formula from Krylov [19] are not checked. These bounds are load-bearing: they imply the explosion estimate (3.13), the tail estimates in Lemma 4.2, and the averaged H1 bound (4.17) used for tightness. The proof needs a regularization argument (e.g., Galerkin or mollification) or a precise verification of the hypotheses of the cited Itô formula.
  2. [§4.1, Lemma 4.1] The displayed Itô formula for d/dt E∥u(t)∥^p_{L2} omits the nonnegative second-order term (1/2)p(p-2)E[∥u∥^{p-4}_{L2} Σ_j (∫_R u σ(u) a_j e_j dx)^2]. Hence the equality preceding (4.4) is false. The final inequality (4.1) can be recovered: by Cauchy-Schwarz and (H1), the omitted term is bounded by the same p(p-1)/2 a l^2 E∥u∥^p used in the text. But the proof as written is incorrect and should be rewritten as an inequality from the start.
  3. [§4.3, Proposition 4.4] The proof asserts that any φ ∈ C_b(L2) is uniformly continuous on the ball B(0,N) in L2(R). This is false because B(0,N) is not compact in infinite dimensions. Since the Feller property is required for the Krylov-Bogolioubov theorem, the proof is incomplete. The gap is repairable: one can prove pointwise continuity by combining Lemma 4.3 with convergence in probability and dominated convergence, rather than uniform continuity on balls. As written, the step around (4.12) is a genuine error.
  4. [§3.1, Lemma 3.3] The proof of the maximal inequality for the stochastic convolution contains an unjustified step after Burkholder's inequality: the displayed chain changes the order of the Lp norm and the time integral, and the exponent is altered (p/2 inside the spatial integral versus 2 outside), so the claimed bound E sup_t ∥Gφ(t)∥^q_{Lp} ≤ C E∫_0^T ∥φ(s)∥^q_{Lp} ds does not follow from the displayed inequalities. Since this estimate is used to control the stochastic term A4 in the contraction mapping, a correct proof or a precise citation is needed.
minor comments (5)
  1. [§4.2, Lemma 4.2] After multiplying by e^{(2k-al^2)t} and integrating, the term involving ∥u_x(s)∥^2 should be inside an expectation; the displayed inequality omits the expectation operator.
  2. [§4.3, Equation (4.19)] The definition of Y_m appears to contain a typo: the H1 threshold should presumably involve 2^m sqrt(3c1c2/ε) rather than 2m sqrt(3c1c2)√ε, given the probability bound on the following lines.
  3. [§4.3, Proposition 4.4] Equation (3.13) is invoked for p=2, but (3.13) was derived for the truncated solution u_N; the notation should clarify that the same estimate holds for the stopped solution used in Proposition 4.4.
  4. [General] There are several language and grammar issues, such as "Specially" for "In particular" and "making it difficult to derive" in Section 1; these should be corrected in a revision.
  5. [§4.1, Lemma 4.1] The paper cites Krylov [19] for Itô's formula, but the statement of Lemma 4.1 does not mention the regularity needed to apply it; a sentence connecting the hypotheses to the cited result would help the reader.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the invariant measure proof is a standard self-contained derivation via existence theory, uniform estimates, tightness, and the Krylov–Bogolioubov theorem.

full rationale

The paper's derivation chain is not circular. Theorem 2.6 is obtained by combining: (i) the existence and uniqueness result Theorem 2.3, proved by truncation and fixed-point arguments; (ii) uniform-in-time bounds Lemma 4.1, obtained from Itô's formula; (iii) uniform tail estimates Lemma 4.2, obtained by a cut-off function and Agmon's inequality; (iv) tightness of the empirical laws via compact sets; and (v) the Feller property proved in Proposition 4.4 through a stopping-time argument and Lemma 4.3. None of these steps assumes the existence of an invariant measure, and no fitted parameter is renamed as a prediction. The paper cites external results such as Krylov's Itô formula and heat-kernel estimates, but these are standard tools whose hypotheses are stated, and the paper does not rely on self-citations to close an argument. The proof does contain nontrivial justification gaps: Lemma 3.5 explicitly says the Itô calculations 'are formal and can be justified by a limiting procedure,' and Proposition 4.4 appears to invoke uniform continuity of an arbitrary bounded continuous function on the noncompact ball B(0,N). These are correctness and regularity concerns, not circularity, because the conclusion is not built into an input or derived from a self-citation. Accordingly, the circularity score is 0.

Assumptions & free parameters 0 free parameters · 4 assumptions · 0 invented entities

The paper contributes no new free parameters or invented entities. It relies on standard SPDE tools and the technical condition a l^2 < 3k/7, which is an assumption on the given coefficients, not a fitted parameter.

assumptions (4)
  • standard math Ito's formula for L^p norms of stochastic W^1_p-valued processes (Krylov [19]) is applicable to the mild solution.
    Used to derive the energy estimates in Lemmas 3.5 and 4.1; the paper says the calculations are formal and can be justified by a limiting procedure, but does not carry it out.
  • standard math Heat kernel operator estimates J1 and J2 from Gyongy-Nualart [15] (Lemmas 3.1 and 3.2) hold as quoted.
    These estimates are the basis for the local contraction argument.
  • standard math Krylov-Bogolioubov theorem for Feller semigroups on Polish spaces can be applied to L2(R).
    Used to convert tightness plus the Feller property into existence of an invariant measure.
  • standard math Agmon's inequality in one dimension, ||u||_{L infinity} <= C ||u||_{L2}^{1/2} ||u_x||_{L2}^{1/2}.
    Used in Lemma 4.2 to bound the L^3 norm in the tail estimates.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Invariant measures for stochastic Burgers equation on unbounded domains." pith.science (2026). https://pith.science/paper/VSI2XJZW

@misc{pith2026250607119,
  author       = {Pith},
  title        = {Pith review of: Invariant measures for stochastic Burgers equation on unbounded domains},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/VSI2XJZW}},
  note         = {Machine review of arXiv:2506.07119}
}
read the original abstract

In this paper, we investigate the stochastic damped Burgers equation with multiplicative noise defined on the entire real line. We demonstrate the existence and uniqueness of a mild solution to the stochastic damped Burgers equation and establish that the solution is uniformly bounded in time. Furthermore, by employing the uniform estimates on the tails of the solution, we obtain the tightness of a family of probability distributions of the solution. Subsequently, by applying the Krylov-Bogolioubov theorem, we establish the existence of invariant measures.

Discussion (0). Sign in to comment.

Reference graph

Works this paper leans on

22 extracted references · 22 canonical work pages

  1. [19]

    N. V. Krylov, Itˆ o’s formula for theL p-norm of stochasticW 1 p -valued processes.Probab. Theory Related Fields147(2010), 583–605

    work page 2010

  2. [1]

    Assing and R

    S. Assing and R. Manthey, Invariant measures for stochastic heat equations with unbounded coefficients. Stochastic Process. Appl.103(2003), 237–256

    work page 2003

  3. [2]

    Bakhtin and L

    Y. Bakhtin and L. Li, Thermodynamic limit for directed polymers and stationary solutions of the Burgers equation.Comm. Pure Appl. Math.72(2019), 536–619

    work page 2019

  4. [3]

    Bertini, N

    L. Bertini, N. Cancrini and G. Jona-Lasinio, The stochastic Burgers equation.Comm. Math. Phys.165 (1994), 211–232

    work page 1994

  5. [4]

    Brze´ zniak, E

    Z. Brze´ zniak, E. Motyl and M. Ondrejat, Invariant measure for the stochastic Navier-Stokes equations in unbounded 2D domains.Ann. Probab.45(2017), 3145–3201

    work page 2017

  6. [5]

    Brze´ zniak, M

    Z. Brze´ zniak, M. Ondrej´ at and J. Seidler, Invariant measures for stochastic nonlinear beam and wave equations.J. Differential Equations260(2016), 4157–4179

    work page 2016

  7. [6]

    Da Prato, A

    G. Da Prato, A. Debussche and R. Temam, Stochastic Burgers’ equation.NoDEA Nonlinear Differential Equations Appl.1(1994), 389–402. 20 ZHENXIN LIU AND ZHIYUAN SHI

    work page 1994

  8. [7]

    Da Prato and D

    G. Da Prato and D. G¸ atarek, Stochastic Burgers equation with correlated noise.Stochastics Stochastics Rep.52(1995), 29–41

    work page 1995

Show all 22 references
  1. [8]

    Dong and R

    Z. Dong and R. Zhang, Ergodicity for a class of semilinear stochastic partial differential equations.Math. Methods Appl. Sci.43(2020), 2117–2136

    2020

  2. [9]

    Dunlap, C

    A. Dunlap, C. Graham and L. Ryzhik, Stationary solutions to the stochastic Burgers equation on the line.Comm. Math. Phys.382(2021), 875–949

    2021

  3. [10]

    W. E, K. Khanin, A. Mazel and Ya. Sinai, Invariant measures for Burgers equation with stochastic forcing.Ann. of Math. (2)151(2000), 877–960

    2000

  4. [11]

    Eckmann and M

    J.-P. Eckmann and M. Hairer, Invariant measures for stochastic partial differential equations in un- bounded domains.Nonlinearity14(2001), 133–151

    2001

  5. [12]

    Friedman,Partial Differential Equations of Parabolic Type, Prentice-Hall

    A. Friedman,Partial Differential Equations of Parabolic Type, Prentice-Hall. Inc., Englewood Cliffs, NJ, 1964

    1964

  6. [13]

    Goldys and B

    B. Goldys and B. Maslowski, Exponential ergodicity for stochastic Burgers and 2D Navier-Stokes equa- tions.J. Funct. Anal.226(2005), 230–255

    2005

  7. [14]

    Gy¨ ongy, Existence and uniqueness results for semilinear stochastic partial differential equations.Sto- chastic Process

    I. Gy¨ ongy, Existence and uniqueness results for semilinear stochastic partial differential equations.Sto- chastic Process. Appl.73(1998), 271–299

    1998

  8. [15]

    Gy¨ ongy and D

    I. Gy¨ ongy and D. Nualart, On the stochastic Burgers’ equation in the real line.Ann. Probab.27(1999), 782–802

    1999

  9. [16]

    Gy¨ ongy and C

    I. Gy¨ ongy and C. Rovira, OnL p-solutions of semilinear stochastic partial differential equations.Sto- chastic Process. Appl.90(2000), 83–108

    2000

  10. [17]

    J. U. Kim, Invariant measures for a stochastic nonlinear Schr¨ odinger equation.Indiana Univ. Math. J. 55(2006), 687–717

    2006

  11. [18]

    J. U. Kim, On the stochastic Burgers equation with a polynomial nonlinearity in the real line.Discrete Contin. Dyn. Syst. Ser. B6(2006), 835–866

    2006

  12. [20]

    Misiats, O

    O. Misiats, O. Stanzhytskyi and N. K. Yip, Existence and uniqueness of invariant measures for stochastic reaction-diffusion equations in unbounded domains.J. Theoret. Probab.29(2016), 996–1026

    2016

  13. [21]

    Tessitore and J

    G. Tessitore and J. Zabczyk, Invariant measures for stochastic heat equations.Probab. Math. Statist.18 (1998), 271–287

    1998

  14. [22]

    Wang, Dynamics of fractional stochastic reaction-diffusion equations on unbounded domains driven by nonlinear noise.J

    B. Wang, Dynamics of fractional stochastic reaction-diffusion equations on unbounded domains driven by nonlinear noise.J. Differential Equations268(2019), 1–59. Z. Liu: School of Mathematical Sciences, Dalian University of Technology, Dalian 116024, P. R. China Email address:z...

    2019

Pith tools

Reviewed August 7, 2026 · model on record in the stance chip above.