Pith. sign in

REVIEW 3 major objections 5 minor 62 references

Averaging Principle and Pullback Attractor Convergence for McKean--Vlasov Stochastic Reaction--Diffusion Equations

T0 review · 3 major / 5 minor · reviewed 2026-08-11 · deepseek-v4-flash

Pith's one-line read This paper proves that rapidly oscillating, distribution-dependent stochastic reaction–diffusion equations on the torus are governed, in the small-oscillation limit, by a single averaged equation; the approximation holds on finite…

desk verdict A serious extension of averaging principles to McKean–Vlasov SPDEs, but the finite-time proof has an unjustified double limit that takes down the main theorems. read the letter →

arxiv 2608.09319 v1 pith:CRSVYWVU submitted 2026-08-10 math.DS math.PR

classification math.DSmath.PR MSC 60H1570K6537L3037L55
keywords McKean–VlasovSPDEsstochasticreaction–diffusionequationsaveragingprincipleWassersteindistancepullbackattractorconvergenceBogolyubovtheoremmean-fieldinteraction
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 establishes three averaging principles for McKean–Vlasov stochastic reaction–diffusion equations on the torus with rapidly oscillating coefficients, namely equations whose drift and noise depend on the solution's own law. Theorems 2.3–2.5 show that, as the oscillation period tends to zero, solutions converge in mean square to those of an averaged equation, uniformly on finite time intervals and, under a contraction condition, uniformly over the whole real line. At the level of probability laws, the pullback attractors of the nonautonomous original system converge upper-semicontinuously to the global attractor of the averaged system, uniformly over all time translates of the coefficients. A sympathetic reader should care because this moves averaging theory from finite-time statements for distribution-dependent SPDEs to global, attractor-level statements, and provides a class of mean-field feedback models, including a neural-field-like example, where the results apply.

What carries the argument

The argument is carried by a block-averaging, or freezing, scheme for the mild formulation of (1.1). The rapidly oscillating functions are compared against step processes on small time blocks, with errors controlled by the uniform ergodicity rates $\omega_f$, $\omega_g$, $\omega_\gamma$ of (H3), by the analytic semigroup estimates of Lemma 3.4, and by the discretization estimate of Lemma 3.5 for the cubic term. The whole-line result uses the exponential contraction $\rho = 2(\lambda_1+\lambda) - 4L_f - 4L_g^2 > 0$ to compare both equations on backward intervals $[t-N,t]$ and then let $N\to\infty$. The attractor result uses, in addition, the cocycle structure on the coefficient hull, a uniform second-moment tail estimate yielding relative compactness in $(\mathcal P_2(H), W_2)$, and the density of $\mathcal P_4(V)$ in $\mathcal P_2(H)$, which converts finite-time averaging for smooth initial laws into uniform averaging on compact sets of initial laws.

What would settle it

Compute the size of the constant $C_{T,R}$ produced by the singular Gronwall estimate (Lemma A.6) when applied to the kernel $(t-s)^{-1/2}$ in the proof of Theorem 2.3; if $C_{T,R}$ grows faster than quadratically in $R$, the term $C_{T,R}R^{-2}$ diverges as $R\to\infty$ and the double limit $\varepsilon\to 0$, $R\to\infty$ is unjustified, which would invalidate Theorems 2.4 and 2.5 as proved.

Watch

Extended reading notes

Core claim

The central claim is that the fast time oscillations in the coefficients of equation (1.1), including the coefficient $\gamma(t/\varepsilon)$ of the non-Lipschitz cubic term, can be averaged out: the solution $u^\varepsilon(t)$ is close in mean square to the solution $\bar u(t)$ of (1.2) with averaged coefficients $\bar\gamma$, $\bar f$, $\bar g$. Theorem 2.3 proves this uniformly on $[0,T]$ for initial data with four finite moments in the space $V$. Theorem 2.4 adds the contraction condition $\lambda_1+\lambda > 2L_f + 2L_g^2$, under which both equations possess unique bounded entire solutions and $\sup_{t\in\mathbb R}\mathbb E\|u^\varepsilon(t)-\bar u(t)\|^2\to 0$. Theorem 2.5 lifts the result to the space $(\mathcal P_2(H), W_2)$ of probability laws: the pullback attractors $A^\varepsilon(F)$ of the original cocycle, one for each coefficient symbol $F$ in the hull (the closure of the time translates of the coefficients), converge upper-semicontinuously to the global attractor $\bar A$ of the averaged semigroup, uniformly in the symbol. The work thereby extends the Bogolyubov averaging principle from finite-dimensional and finite-time settings to global, law-valued asymptotics for distribution-dependent SPDEs.

Load-bearing premise

The whole chain of results rests on the limiting step after equation (3.40), where the proof sends the oscillation scale to zero and then the cut-off level to infinity, while the error constant grows with the cut-off level.

Editorial extensions

If this is right

  • For any fixed time horizon $T$, the mean-square difference between the original and averaged solutions with the same initial condition tends to zero uniformly in $t\in[0,T]$.
  • Under the contraction condition, the unique bounded entire solutions of the two equations are indistinguishable in mean square uniformly over all $t\in\mathbb R$, so the averaged equation captures the whole-line statistical dynamics.
  • The long-time law dynamics of the original nonautonomous system, encoded by its pullback attractors, collapse onto the global attractor of the averaged system as $\varepsilon\to 0$, uniformly over the hull of coefficient translations.
  • For the explicit almost-periodic mean-field feedback model of Example 2.6, all three conclusions hold, so there is a concrete family of neural-field-like interacting systems to which the theory applies.

Reading between the lines

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

  • Extending beyond the paper, the block-averaging rates visible in the estimates (an $\varepsilon^{1/8}$ term from the cutoff and block size, together with $\omega(\varepsilon^{-1/2})$ terms) suggest that quantitative, explicit rates could be derived for finite-time and whole-line convergence in this class, though the paper itself does not isolate such rates.
  • Extending beyond the paper, near the threshold $\lambda_1+\lambda = 2L_f + 2L_g^2$ the exponential contraction degenerates; the whole-line and attractor arguments would likely need a polynomial-in-time stability estimate, and the same averaging conclusion may or may not survive.
  • Extending beyond the paper, the same combination of a sign-definite cubic term, Lipschitz-in-law coefficients, and uniform ergodicity of the fast coefficients could be applied to other locally monotone distribution-dependent SPDEs, such as stochastic Allen–Cahn equations with mean-field coupling, but this is not established here.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

3 major / 5 minor

Summary. The paper claims three averaging results for a McKean–Vlasov stochastic reaction--diffusion equation with rapidly oscillating coefficients on the torus, d ≤ 3: (i) finite-time mean-square convergence to the averaged equation (Theorem 2.3), (ii) whole-line mean-square convergence of unique bounded entire solutions under a contraction condition (Theorem 2.4), and (iii) upper-semicontinuous convergence of pullback attractors in the Wasserstein space P2(H) (Theorem 2.5). The proofs are built on a priori estimates in H and V, a time-discretization/stopping-time argument, a fractional Gronwall lemma, and a uniform second-moment tail estimate for compactness in P2(H). An illustrative mean-field feedback model is given in Section 2.3.

Significance. If the results were correct, they would constitute a substantial advance: they appear to be the first finite-time and whole-line averaging principles for distribution-dependent SPDEs with cubic nonlinearity, and the first attractor-convergence result for the law dynamics in Wasserstein space over a noncompact coefficient hull. The paper also contains several genuinely useful technical ingredients, including the density of P4(V) in P2(H) (Lemma 4.11) and the Wasserstein compactness criterion (Lemma 4.6). However, the central finite-time theorem contains a load-bearing gap in its final limiting argument, and since Theorems 2.4 and 2.5 invoke Theorem 2.3, the significance is conditional on a repair that is not a minor edit.

major comments (3)
  1. [Section 3.2, Eqs. (3.39)–(3.40) and Lemma A.6] The claimed double limit in the proof of Theorem 2.3 is not justified. After combining (3.39) and (3.40), the proof obtains y(t) ≤ A_{ε,R,T} + C_{R,T}∫_0^t (t-s)^{-1/2} y(s) ds, with A_{ε,R,T} = C_{T,R}[ε^{1/8} + ω_f(ε^{-1/2})^2 + ω_γ(ε^{-1/2})^2 + ω_g(ε^{-1/2}) + R^{-2}]. Lemma A.6 then yields a bound of the form C_{T,R} times the same bracket. The constant C_{R,T} is inherited from the stopping-time estimates: on {t ≤ τ_R} one only controls ∥uε(s)∥_{H^1} + ∥ū(s)∥_{H^1} ≤ 2R, so the kernel constant in (3.34) is proportional to R^2. Lemma A.6 therefore produces a prefactor E_{1/2}(C R^2 Γ(1/2) T^{1/2}), and since E_{1/2}(z) grows like exp(C z^2) for large real z, the term C_{T,R} R^{-2} diverges as R→∞. Thus the displayed 'Letting ε→0 and then R→∞' is invalid: with R fixed the R^{-2} term does not vanish as ε→0, and with R depending on ε the prefactor grows faster than any polynomial, so the product still fails to vanish. Theorem 2.3 is therefore not proved as written.
  2. [Theorems 2.4 and 2.5] The subsequent main results inherit the gap. The proof of Theorem 2.4 uses (3.43), which is exactly the finite-time estimate of Theorem 2.3 applied on the shifted interval, and the proof of Theorem 2.5 uses Lemma 4.12, which invokes Theorem 2.3. Since the finite-time convergence is not established, the whole-line convergence (2.8) and the attractor convergence in Theorem 2.5 are unsupported.
  3. [Remark 3.6(i)] Remark 3.6(i) claims that the finite-time convergence is uniform over initial distributions with a common V-fourth moment bound. This uniformity is obtained by the same R→∞ argument that is invalid in Theorem 2.3. As written, the assertion does not follow from the preceding estimates, because the uncontrolled dependence of C_{T,R} on R prevents the required uniform limit.
minor comments (5)
  1. [Equation (3.22)] The notation I_{12}^2 after (3.22) is ambiguous: it denotes the third summand in the preceding display, not the square of I_{12}(t). Please rename this term.
  2. [Lemma 3.5] The constant C_{T,R} in Lemma 3.5 is not quantified. Since the main proof depends on the growth of this constant in R, the lemma should either state the dependence explicitly or the proof should avoid relying on it in the final double limit.
  3. [Equation (3.39)] In the passage from (3.39) to the following line, the additive +1 inside the integral is dropped without comment. This is harmless for t ≤ T but should be stated for clarity.
  4. [Lemma A.6] Lemma A.6 does not record the dependence of C_{T,β,B} on B. Given that this dependence is crucial in the proof of Theorem 2.3, it would be helpful to state the Mittag–Leffler bound explicitly, including the exponential growth of E_{1/2} for large arguments.
  5. [Throughout] There are several typographical and formatting issues, including the title 'A VERAGING PRINCIPLE' and some garbled equation displays in Section 2.1; these should be corrected in a revision.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity: the main results are derived in-paper from independent hypotheses; averaging coefficients are defined by time-means, not by solution limits, and self-citations are used only for comparison.

full rationale

The paper is essentially self-contained and its derivation chain does not reduce to its inputs. The averaged coefficients in (H3) are defined by time averages of the given functions f, g, and gamma, independently of the solutions u^epsilon and bar-u; the finite-time theorem then proves convergence of the solutions to the solution of the equation built from those independent averages. The proof of Theorem 2.3 decomposes the difference u^epsilon - bar-u into I_1, I_2, and M, and each term is bounded using the standing hypotheses (H1)-(H3) together with in-paper lemmas (Lemmas 3.1, 3.2, 3.3, 3.5) and standard semigroup/heat-kernel estimates. The final Gronwall-type application (Lemma A.6) turns the integral inequality into an explicit bound; this is a standard mathematical step, not a definitional equivalence. Theorems 2.4 and 2.5 invoke Theorem 2.3, but Theorem 2.3 is proved rather than assumed, so the later results build on an internally established statement rather than on a fitted or self-referential input. Self-citations [15,17] appear only in the introduction as comparisons ('Compared with [15,17]') and are not load-bearing for any proof; the auxiliary results in Appendix A (Lemmas A.2, A.3, A.4, A.5) are proved inside the paper, while external references [20,39,43,56] provide standard existence, semigroup, and compactness tools that do not contain the paper's conclusions. The displayed limiting argument 'Letting epsilon -> 0 and then R -> infinity' after inequality (3.40) may be a genuine correctness gap because the constant C_{T,R} produced by Lemma A.6 with B = C_{R,T} is not shown to be compatible with the R^{-2} term; however, a missing estimate or an unjustified interchange of limits is a correctness issue, not circularity. It does not make the theorem equivalent to its assumptions, and no quantity is fitted, renamed, or defined in terms of the claimed conclusion. Therefore the appropriate circularity score is 0.

Assumptions & free parameters 0 free parameters · 3 assumptions · 0 invented entities

The paper introduces no free parameters fitted to data and no new physical or mathematical entities. All input is contained in the explicit structural assumptions (H1)-(H4) and standard stochastic analysis tools, which are reasonable for this problem class.

assumptions (3)
  • standard math Ito formula, stochastic convolution maximal inequality, analytic semigroup estimates for A = Delta - lambda I, Poincare inequality on the zero-mean subspace, compact embedding H^1(T^d) into L^2(T^d), and Villani's compactness criterion in P2(H).
    Used throughout Section 3 and 4; these are classical results invoked without proof.
  • domain assumption Standing assumptions (H1)-(H4): Lipschitz continuity of f and g in state and law, vanishing mean-oscillation condition on f, g, gamma, and the dissipativity inequality.
    These are the explicit structural hypotheses of the model; they define the class of equations covered and are not derived from anything else.
  • domain assumption The dimension restriction d <= 3 and the zero-mean functional setting H and V defined in (1.4).
    Used for the Sobolev embeddings and integrability of the heat-kernel singularity in Lemma 3.5 and the proof of Theorem 2.3.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Averaging Principle and Pullback Attractor Convergence for McKean--Vlasov Stochastic Reaction--Diffusion Equations." pith.science (2026). https://pith.science/paper/CRSVYWVU

@misc{pith2026260809319,
  author       = {Pith},
  title        = {Pith review of: Averaging Principle and Pullback Attractor Convergence for McKean--Vlasov Stochastic Reaction--Diffusion Equations},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/CRSVYWVU}},
  note         = {Machine review of arXiv:2608.09319}
}
abstract

We establish three averaging principles for distribution-dependent stochastic reaction--diffusion equations with rapidly oscillating coefficients on the torus $\mathbb T^d$, $d\le3$. First, solutions converge in mean square, uniformly on finite time intervals, to solutions of the averaged equation. Under a contraction condition, both the original and averaged equations admit unique bounded entire solutions whose mean-square distance vanishes uniformly for all $t\in\mathbb R$. At the level of probability laws, the original nonautonomous equation possesses a family of pullback attractors, whereas the averaged equation has a global attractor; the former converge upper-semicontinuously to the latter, uniformly over the coefficient hull. As an application, we present a class of stochastic reaction--diffusion models motivated by large-scale interacting systems.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

62 extracted references · 61 canonical work pages

  1. [1]

    Bakhtin and Y

    V. Bakhtin and Y. Kifer. Diffusion approximation for slow motion in fully coupled averaging.Probab. Theory Related Fields, 129(2):157–181, 2004

  2. [2]

    J. Bao, G. Yin and C. Yuan. Two-time-scale stochastic partial differential equations driven byα-stable noises: averaging principles.Bernoulli, 23(1):645–669, 2017

  3. [3]

    N. N. Bogolyubov and Y. A. Mitropolsky.Asymptotic Methods in the Theory of Non-linear Oscillations, volume 10. CRC Press, 1961

  4. [4]

    Br´ ehier

    C.-E. Br´ ehier. Strong and weak orders in averaging for SPDEs.Stochastic Process. Appl., 122:2553–2593, 2012

  5. [5]

    P. C. Bressloff. Spatiotemporal dynamics of continuum neural fields.J. Phys. A, 45(3):033001, 2012

  6. [6]

    Buckdahn, J

    R. Buckdahn, J. Li, S. Peng, and C. Rainer. Mean-field stochastic differential equations and associated PDEs. Ann. Probab., 45(2):824–878, 2017

  7. [7]

    Caraballo, G

    T. Caraballo, G. Lukaszewicz and J. Real. Pullback attractors for asymptotically compact non-autonomous dynamical systems.Nonlinear Anal., 64(3):484–498, 2006

  8. [8]

    Carmona and F

    R. Carmona and F. Delarue.Probabilistic Theory of Mean Field Games with Applications I–II. Volumes 83–84 ofProbab. Theory Stoch. Model.. Springer, Cham, 2018

Show all 62 references
  1. [9]

    S. Cerrai. A Khasminskii type averaging principle for stochastic reaction–diffusion equations.Ann. Appl. Probab., 19(3):899–948, 2009

  2. [10]

    S. Cerrai. Averaging principle for systems of reaction-diffusion equations with polynomial nonlinearities perturbed by multiplicative noise.SIAM J. Math. Anal., 43(6):2482–2518, 2011

  3. [11]

    Cerrai and M

    S. Cerrai and M. I. Freidlin. Averaging principle for a class of stochastic reaction–diffusion equations.Probab. Theory Related Fields, 144(1):137–177, 2009

  4. [12]

    Cerrai and A

    S. Cerrai and A. Lunardi. Averaging principle for nonautonomous slow-fast systems of stochastic reaction- diffusion equations: the almost periodic case.SIAM J. Math. Anal., 49(4):2843–2884, 2017

  5. [13]

    Cheban and Z

    D. Cheban and Z. Liu. Averaging principle on infinite intervals for stochastic ordinary differential equations. Electron. Res. Arch., 29(4):2791–2817, 2021

  6. [14]

    Cheng, Z

    M. Cheng, Z. Hao, and M. R¨ ockner. Strong and weak convergence for the averaging principle of DDSDE with singular drift.Bernoulli, 30(2):1586–1610, 2024

  7. [15]

    Cheng and Z

    M. Cheng and Z. Liu. The second bogolyubov theorem and global averaging principle for SPDEs with monotone coefficients.SIAM J. Math. Anal., 55(2):1100–1144, 2023

  8. [16]

    Cheng, Z

    M. Cheng, Z. Liu, and M. R¨ ockner. Averaging principle and normal deviation for multi-scale SDEs with polynomial nonlinearity.arXiv preprint arXiv:2308.10751, 2023. To appear inAnn. Appl. Probab

  9. [17]

    Cheng, Z

    M. Cheng, Z. Liu, and M. R¨ ockner. Averaging principle for stochastic complex Ginzburg-Landau equations. J. Differential Equations, 368:58–104, 2023

  10. [18]

    Cheng, X

    M. Cheng, X. Sun, and Y. Xie. Strong averaging principle for nonautonomous multi-scale SPDEs with fully local monotone and almost periodic coefficients.J. Funct. Anal., 290(10):Paper No. 111411, 43, 2026. A VERAGING PRINCIPLE AND PULLBACK ATTRACTOR CONVERGENCE 35

  11. [19]

    Crisan, P

    D. Crisan, P. Dobson, B. D. Goddard, M. Ottobre, and I. Souttar. Poisson equations with locally-Lipschitz coefficients and uniform in time averaging for stochastic differential equations via strong exponential stability. Ann. Inst. Henri Poincar´ e Probab. Stat., 62(1):667–721, 2026

  12. [20]

    Da Prato and J

    G. Da Prato and J. Zabczyk.Stochastic Equations in Infinite Dimensions. Encyclopedia Math. Appl., 44. Cambridge University Press, Cambridge, 1992

  13. [21]

    F. de Feo. The order of convergence in the averaging principle for slow-fast systems of stochastic evolution equations in Hilbert spaces.Appl. Math. Optim., 88(2):Paper No. 39, 2023

  14. [22]

    Z. Dong, X. Sun, H. Xiao, and J. Zhai. Averaging principle for one dimensional stochastic Burgers equation. J. Differential Equations, 265(10):4749–4797, 2018

  15. [23]

    Duan and W

    J. Duan and W. Wang.Effective Dynamics of Stochastic Partial Differential Equations. Elsevier, Amsterdam, 2014

  16. [24]

    G. B. Ermentrout and D. H. Terman.Mathematical Foundations of Neuroscience, volume 35. Interdiscip. Appl. Math., 35. Springer, New York, 2010

  17. [25]

    M. I. Freidlin and A. D. Wentzell.Random Perturbations of Dynamical Systems, volume 260 ofGrundlehren Math. Wiss.[Fundamental Principles of Mathematical Sciences]. Springer-Verlag, New York, second edition,

  18. [26]

    M. I. Freidlin and A. D. Wentzell. Long-time behavior of weakly coupled oscillators.J. Stat. Phys., 123(6):1311–1337, 2006

  19. [27]

    H. Fu, L. Wan, and J. Liu. Strong convergence in averaging principle for stochastic hyperbolic-parabolic equations with two time-scales.Stochastic Process. Appl., 125(8):3255–3279, 2015

  20. [28]

    P. Gao. Bogoliubov averaging principle of stochastic reaction–diffusion equation.J. Differential Equations, 266(10):6702–6738, 2019

  21. [29]

    Hairer and X.-M

    M. Hairer and X.-M. Li. Averaging dynamics driven by fractional Brownian motion.Ann. Probab., 48(4):1826– 1860, 2020

  22. [30]

    W. Hong, S. Li, and W. Liu. Strong convergence rates in averaging principle for slow-fast McKean–Vlasov SPDEs.J. Differential Equations, 316:94–135, 2022

  23. [31]

    Kamenskii, O

    M. Kamenskii, O. Mellah, and P. Raynaud de Fitte. Weak averaging of semilinear stochastic differential equations with almost periodic coefficients.J. Math. Anal. Appl., 427(1):336–364, 2015

  24. [32]

    R. Z. Khasminskii. On stochastic processes defined by differential equations with a small parameter.Theor. Probab. Appl., 11(2):211–228, 1966

  25. [33]

    R. Z. Khasminskii. On the principle of averaging the Itˆ o’s stochastic differential equations.Kybernetika, 4(3):260–279, 1968

  26. [34]

    R. Z. Khasminskii and G. Yin. On averaging principles: an asymptotic expansion approach.SIAM J. Math. Anal., 35(6):1534–1560, 2004

  27. [35]

    Y. Kifer. Some recent advances in averaging.Modern dynamical systems and applications, 385–403. Cambridge University Press, Cambridge, 2004

  28. [36]

    N. M. Krylov and N. N. Bogolyubov.Introduction to Non-linear Mechanics.Ann. of Math. Stud., 11. Prince- ton University Press, Princeton, NJ, 1943

  29. [37]

    Lasry and P.-L

    J.-M. Lasry and P.-L. Lions. Mean field games.Jpn. J. Math., 2(1):229–260, 2007

  30. [38]

    Y. Li, F. Wu, and L. Xie. Poisson equation on Wasserstein space and diffusion approximations for multiscale McKean-Vlasov equation.SIAM J. Math. Anal., 56(2):1495–1524, 2024

  31. [39]

    Liu and M

    W. Liu and M. R¨ ockner.Stochastic Partial Differential Equations: an Introduction. Universitext. Springer, Cham, 2015

  32. [40]

    W. Liu, M. R¨ ockner, X. Sun, and Y. Xie. Averaging principle for slow-fast stochastic differential equations with time dependent locally lipschitz coefficients.J. Differential Equations, 268(6):2910–2948, 2020

  33. [41]

    Liu and Y

    X. Liu and Y. Wang. Averaging principle on infinite intervals for stochastic ordinary differential equations with L´ evy noise.AIMS Math., 6(5):5316–5350, 2021

  34. [42]

    H. P. McKean, Jr. A class of Markov processes associated with nonlinear parabolic equations.Proc. Nat. Acad. Sci. U.S.A., 56(6):1907–1911, 1966

  35. [43]

    Pazy.Semigroups of Linear Operators and Applications to Partial Differential Equations

    A. Pazy.Semigroups of Linear Operators and Applications to Partial Differential Equations. Appl. Math. Sci., 44. Springer-Verlag, New York, 1983

  36. [44]

    B. Pei, Y. Xu, and G. Yin. Stochastic averaging for a class of two-time-scale systems of stochastic partial differential equations.Nonlinear Anal., 160:159–176, 2017

  37. [45]

    B. Pei, Y. Inahama, and Y. Xu. Averaging principle for fast-slow system driven by mixed fractional Brownian rough path.J. Differential Equations, 301:202–235, 2021

  38. [46]

    R¨ ockner, X

    M. R¨ ockner, X. Sun, and Y. Xie. Strong convergence order for slow-fast McKean–Vlasov stochastic differential equations.Ann. Inst. Henri Poincar´ e Probab. Stat., 57(1):547–576, 2021

  39. [47]

    R¨ ockner, L

    M. R¨ ockner, L. Xie, and L. Yang. Asymptotic behavior of multiscale stochastic partial differential equations with H¨ older coefficients.J. Funct. Anal., 285(9):Paper No. 110103, 50, 2023. 36 HONGLEI CHEN, MENGYU CHENG, AND ZHENXIN LIU

  40. [48]

    G. Shen, J. Xiang, and J.-L. Wu. Averaging principle for distribution dependent stochastic differential equations driven by fractional Brownian motion and standard Brownian motion.J. Differential Equations, 321:381–414, 2022

  41. [49]

    G. Shen, J. Yin, and J.-L. Wu. Stochastic averaging principle for two-time-scale SDEs with distribution- dependent coefficients driven by fractional Brownian motion.Commun. Math. Stat., 13(6):1445–1479, 2025

  42. [50]

    Y. Shi, H. Liu, and H. Gao. Weak and strong averaging principle for non-autonomous slow-fast McKean– Vlasov SDEs with almost periodic coefficients.Math. Methods Appl. Sci., 48(16):15111–15131, 2025

  43. [51]

    X. Sun, L. Xie, and Y. Xie. Averaging principle for slow-fast stochastic partial differential equations with H¨ older continuous coefficients.J. Differential Equations, 270:476–504, 2021

  44. [52]

    Sznitman

    A.-S. Sznitman. Topics in propagation of chaos. In ´Ecole d’ ´Et´ e de Probabilit´ es de Saint-Flour XIX—1989, volume 1464 ofLecture Notes in Math., 165–251. Springer, Berlin, 1991

  45. [53]

    K. Uda. Averaging principle for stochastic differential equations in the random periodic regime.Stochastic Process. Appl., 139:1–36, 2021

  46. [54]

    A. Y. Veretennikov. On an averaging principle for systems of stochastic differential equations.Math. USSR- Sb., 69(1):271–284, 1991

  47. [55]

    full dependence

    A. Y. Veretennikov. On large deviations in the averaging principle for SDEs with a “full dependence”.Ann. Probab., 27(1):284–296, 1999

  48. [56]

    Villani.Optimal Transport: Old and New, volume 338 ofGrundlehren Math

    C. Villani.Optimal Transport: Old and New, volume 338 ofGrundlehren Math. Wiss.[Fundamental Principles of Mathematical Sciences]. Springer, 2009

  49. [57]

    G. Wainrib. Double averaging principle for periodically forced slow-fast stochastic systems.Electron. Com- mun. Probab., 18(51):1–12, 2013

  50. [58]

    Wang and A

    W. Wang and A. J. Roberts. Average and deviation for slow-fast stochastic partial differential equations.J. Differential Equations, 253(5):1265–1286, 2012

  51. [59]

    Xie and X

    L. Xie and X. Zhang. Uniform-in-time diffusion approximations for multiscale stochastic systems.arXiv preprint arXiv:2604.00692, 2026

  52. [60]

    J. Xu, J. Liu, J. Liu, and Y. Miao. Strong averaging principle for two-time-scale stochastic McKean–Vlasov equations.Appl. Math. Optim., 84:S837–S867, 2021

  53. [61]

    W. Yin, Y. Ren, and K. Cao. Stability and averaging principle for distribution dependent SDEs driven by G-Brownian motion.J. Math. Anal. Appl., 555(1):130024, 2026. H. Chen: School of Mathematical Sciences, Dalian University of Technology, Dalian 116024, P. R. China Email addr...

  54. [1998]

    Translated from the 1979 Russian original by Joseph Sz¨ ucs

Pith tools

Reviewed August 11, 2026 · model on record in the stance chip above.