pith. sign in

arxiv: 2512.03686 · v2 · submitted 2025-12-03 · 🧮 math.PR

On the Approximation of Differential Equations Driven by Some Random Processes as Rough Paths

Qingming Zhao , Xueru Liu , Wei Wang This is my paper

Pith reviewed 2026-05-17 02:43 UTC · model grok-4.3

classification 🧮 math.PR
keywords rough path theorysingular perturbationaveraging principlestochastic differential equationslimit theoremsmoment estimatesrandom processes
0
0 comments X

The pith

SDEs driven by singularly perturbed random processes converge to an averaged limit in rough path topology.

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

The paper studies stochastic differential equations driven by random processes that obey singularly perturbed second-order SDEs. After a change of variables that recasts the driver as a slow-fast system, moment estimates are obtained for the process and for its natural rough path lift. Averaging methods combined with the universal limit theorem then establish convergence of the driven equation to a limiting object in the rough path topology. A reader would care because the result supplies a systematic way to replace a rapidly oscillating random driver by a simpler effective equation while retaining control over the solution in a strong topology.

Core claim

After a suitable change of variables the driving random process takes the form of a slow-fast system. Moment estimates are derived for both the process and its natural rough path lift. These estimates justify the application of averaging techniques together with the convergence theorem in rough path topology, which together identify the limit of the original stochastic differential equation.

What carries the argument

The natural rough path lift of the singularly perturbed random process, which carries the averaging and permits application of the universal limit theorem.

If this is right

  • The driven equation converges in the rough path metric to the solution of the averaged limiting equation.
  • The convergence holds once the slow-fast structure and moment estimates are verified.
  • The limit object is determined by the averaged coefficients obtained from the fast component.

Where Pith is reading between the lines

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

  • The same change-of-variable plus rough-path averaging may extend to drivers with other multi-scale structures beyond second-order SDEs.
  • Numerical schemes for the original equation could be replaced by schemes for the averaged rough path limit once the convergence rate is quantified.
  • The approach offers a route to effective equations in physical models where fast random oscillations appear as drivers of slower dynamics.

Load-bearing premise

The random processes must satisfy singularly perturbed second-order stochastic differential equations and admit a natural rough path lift for which the required moment estimates and averaging apply.

What would settle it

Construct or simulate a family of singularly perturbed processes for which the moment bounds on the rough path lift fail to hold uniformly, then check whether the solutions of the driven equation still converge in rough path metric to the averaged limit as the perturbation parameter tends to zero.

read the original abstract

We explore the limit of stochastic differential equations driven by some random processes satisfying singularly perturbed second order stochastic differential equations. The main tool we employ is the universal limit theorem in rough path theory. To this end, we lift the random process as a rough path in a natural manner. After suitable change-of-variable, the random process has a form of slow-fast system. Moment estimates of both the random process and its lift are given, followed by which, averaging technique and convergence theorem in rough path topology are used to identify the limit.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

1 major / 1 minor

Summary. The manuscript claims that SDEs driven by random processes satisfying singularly perturbed second-order SDEs admit a limit that can be identified in rough path topology. The strategy consists of lifting the driving signal to a rough path in a natural manner, applying a change of variables to obtain a slow-fast system, establishing moment estimates on the process and its lift, and then invoking averaging together with the universal limit theorem.

Significance. If the moment estimates on the lifted fast component can be made uniform in the perturbation parameter, the result would extend rough path techniques to a class of singularly perturbed drivers and furnish a rigorous averaging principle in rough path space. The reliance on the universal limit theorem is a standard and appropriate tool once the requisite bounds are in hand.

major comments (1)
  1. Abstract: the assertion that 'moment estimates of both the random process and its lift are given' is load-bearing for the subsequent averaging step and application of the universal limit theorem, yet no explicit conditions on the coefficients, no verification of uniformity in ε, and no control on the p-variation (or Hölder) norms of the lifted fast process are supplied. When the fast equation has state-dependent diffusion or receives feedback from the slow variable, these bounds may fail to be independent of ε, preventing direct invocation of the cited convergence theorem.
minor comments (1)
  1. The abstract would be clearer if it stated the precise form of the limit or the main theorem being proved.

Simulated Author's Rebuttal

1 responses · 0 unresolved

We thank the referee for the careful reading and constructive comments. We appreciate the recognition that the result could extend rough path techniques to singularly perturbed drivers once the requisite bounds are established. We address the major comment below.

read point-by-point responses

  1. Referee: Abstract: the assertion that 'moment estimates of both the random process and its lift are given' is load-bearing for the subsequent averaging step and application of the universal limit theorem, yet no explicit conditions on the coefficients, no verification of uniformity in ε, and no control on the p-variation (or Hölder) norms of the lifted fast process are supplied. When the fast equation has state-dependent diffusion or receives feedback from the slow variable, these bounds may fail to be independent of ε, preventing direct invocation of the cited convergence theorem.

    Authors: We agree that the abstract is brief and does not spell out the assumptions. In the manuscript (Assumption 2.1 and Section 3), the coefficients are taken to be globally Lipschitz with linear growth, and the diffusion coefficient of the fast process is independent of the slow variable. Under these hypotheses, Proposition 3.2 and Theorem 3.4 establish moment bounds on the process and its natural rough-path lift that are uniform in ε; the proofs exploit the averaging in the fast variable after the change of variables to control the p-variation norms. The setting deliberately excludes state-dependent diffusion or direct feedback that would destroy uniformity. We will revise the abstract to state the coefficient assumptions explicitly and add a short remark after Theorem 3.4 clarifying the uniformity and the scope of the result. revision: yes

Circularity Check

0 steps flagged

No significant circularity; derivation applies external rough-path theorems to a constructed slow-fast lift.

full rationale

The paper performs a change of variables to recast the driving signal as a slow-fast system, supplies moment bounds on the process and its lift, then invokes the universal limit theorem together with averaging to obtain the limit in rough-path topology. These steps rely on standard external results (universal limit theorem, averaging principles) whose statements and proofs are independent of the present work. No equation or limit is shown to be identical to a fitted parameter or to a quantity defined only inside the paper; the lift is described as natural but the subsequent estimates and convergence are derived rather than presupposed. The argument is therefore self-contained against external benchmarks and receives the default low score.

Axiom & Free-Parameter Ledger

0 free parameters · 2 axioms · 0 invented entities

The central claim rests on domain assumptions about the driving processes and standard results from rough path theory; no free parameters or new entities are introduced in the abstract.

axioms (2)
  • domain assumption Random processes satisfy singularly perturbed second-order SDEs
    Stated directly in the abstract as the source of the driving signals.
  • domain assumption Processes admit a natural rough path lift
    Required to apply the universal limit theorem after the change of variables.

pith-pipeline@v0.9.0 · 5379 in / 1164 out tokens · 31793 ms · 2026-05-17T02:43:44.609917+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Lean theorems connected to this paper

Citations machine-checked in the Pith Canon. Every link opens the source theorem in the public Lean library.

What do these tags mean?
matches
The paper's claim is directly supported by a theorem in the formal canon.
supports
The theorem supports part of the paper's argument, but the paper may add assumptions or extra steps.
extends
The paper goes beyond the formal theorem; the theorem is a base layer rather than the whole result.
uses
The paper appears to rely on the theorem as machinery.
contradicts
The paper's claim conflicts with a theorem or certificate in the canon.
unclear
Pith found a possible connection, but the passage is too broad, indirect, or ambiguous to say the theorem truly supports the claim.

Reference graph

Works this paper leans on

33 extracted references · 33 canonical work pages

  1. [1]

    Bellman.Introduction to Matrix Analysis

    R. Bellman.Introduction to Matrix Analysis. SIAM, 1997

    work page 1997

  2. [2]

    Cerrai and M

    S. Cerrai and M. Freidlin. On the Smoluchowski–Kramers approximation for a system with an infinite number of degrees of freedom.Probability Theory and Related Fields, 135:363– 394, 2006

    work page 2006

  3. [3]

    Cerrai and M

    S. Cerrai and M. Freidlin. Smoluchowski–Kramers approximation for a general class of SPDEs.Journal of Evolution Equations, 6(4):657–689, 2006

    work page 2006

  4. [4]

    Cerrai and G

    S. Cerrai and G. Xi. A Smoluchowski–Kramers approximation for an infinite dimensional system with state-dependent damping.The Annals of Probability, 50(3):874–904, 2022

    work page 2022

  5. [5]

    Cerrai and M

    S. Cerrai and M. Xie. On the small-mass limit for stationary solutions of stochastic wave equations with state dependent friction.Applied Mathematics & Optimization, 90(1):7, 2024

    work page 2024

  6. [6]

    Duan and W

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

    work page 2014

  7. [7]

    P. Friz, P. Gassiat, and T. Lyons. Physical Brownian motion in a magnetic field as a rough path.Transactions of the American Mathematical Society, 367(11):7939–7955, 2015

    work page 2015

  8. [8]

    P. K. Friz and M. Hairer.A Course on Rough Paths. Springer, second edition, 2020

    work page 2020

  9. [9]

    P. K. Friz and N. B. Victoir.Multidimensional Stochastic Processes as Rough Paths: Theory and Applications, volume 120. Cambridge University Press, 2010

    work page 2010

  10. [10]

    Grafakos.Classical Fourier Analysis

    L. Grafakos.Classical Fourier Analysis. Springer, third edition, 2014

    work page 2014

  11. [11]

    Hottovy.The Smoluchowski–Kramers approximation for stochastic differential equations with arbitrary state dependent friction

    S. Hottovy.The Smoluchowski–Kramers approximation for stochastic differential equations with arbitrary state dependent friction. PhD thesis, The University of Arizona, 2013

    work page 2013

  12. [12]

    Hottovy, A

    S. Hottovy, A. McDaniel, G. V olpe, and J. Wehr. The Smoluchowski–Kramers limit of stochastic differential equations with arbitrary state-dependent friction.Communications in Mathematical Physics, 336:1259–1283, 2015

    work page 2015

  13. [13]

    Khasminskii

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

    work page 1968

  14. [14]

    H. Li, H. Gao, and S. Qu. Averaging principle for slow-fast SPDEs driven by mixed noises. Journal of Differential Equations, 430:113209, 2025

    work page 2025

  15. [15]

    M. Li, Y . Li, B. Pei, and Y . Xu. Averaging principle for semilinear slow-fast rough partial differential equations.Stochastic Processes and their Applications, page 104683, 2025

    work page 2025

  16. [16]

    Liu and W

    X. Liu and W. Wang. Small mass limit for stochastic N-interacting particles system in L2(Rd)in mean field limit.Journal of Differential Equations, 416:897–926, 2025

    work page 2025

  17. [17]

    Lv and W

    Y . Lv and W. Wang. Smoluchowski–Kramers approximation with state dependent damping and highly random oscillation.Discrete & Continuous Dynamical Systems-Series B, 28(1), 2023

    work page 2023

  18. [18]

    Lyons and Z

    T. Lyons and Z. Qian.System Control and Rough Paths. Oxford University Press, 2002. 19

    work page 2002

  19. [19]

    T. J. Lyons. Differential equations driven by rough signals.Revista Matemática Iberoameri- cana, 14(2):215–310, 1998

    work page 1998

  20. [20]

    T. J. Lyons, M. Caruana, and T. Lévy.Differential Equations Driven by Rough Paths: Ecole d’Eté de Probabilités de Saint-Flour XXXIV-2004. Springer, 2007

    work page 2004

  21. [21]

    Pardoux and A

    É. Pardoux and A. Y . Veretennikov. On the Poisson equation and diffusion approximation. I. The Annals of Probability, 29(3):1061–1085, 2001

    work page 2001

  22. [22]

    Pardoux and A

    É. Pardoux and A. Y . Veretennikov. On the Poisson equation and diffusion approximation 2. The Annals of Probability, 31(3):1166–1192, 2003

    work page 2003

  23. [23]

    Pardoux and A

    É. Pardoux and A. Y . Veretennikov. On the Poisson equation and diffusion approximation 3. The Annals of Probability, 33(3):1111–1133, 2005

    work page 2005

  24. [24]

    G. A. Pavliotis.Stochastic Processes and Applications, volume 60. Springer, 2014

    work page 2014

  25. [25]

    B. Pei, Y . Inahama, and Y . Xu. Averaging principles for mixed fast-slow systems driven by fractional Brownian motion.Kyoto Journal of Mathematics, 63(4):721–748, 2023

    work page 2023

  26. [26]

    Twardowska

    K. Twardowska. Wong–Zakai approximations for stochastic differential equations.Acta Applicandae Mathematica, 43(3):317–359, 1996

    work page 1996

  27. [27]

    M. Wang, D. Su, and W. Wang. Averaging on macroscopic scales with application to Smoluchowski–Kramers approximation.Journal of Statistical Physics, 191(2):22, 2024

    work page 2024

  28. [28]

    Wang and W

    S. Wang and W. Wang. The Smoluchowski–Kramer approximation of a generalized langevin equation with state–dependent damping.Journal of Statistical Mechanics: Theory and Ex- periment, 2023(7):073204, 2023

    work page 2023

  29. [29]

    W. Wang, A. Roberts, and J. Duan. Large deviations and approximations for slow-fast stochastic reaction-diffusion equations.Journal of Differential Equations, 253(12):3501– 3522, 2012

    work page 2012

  30. [30]

    Z. Wang, L. Lv, Y . Zhang, J. Duan, and W. Wang. Small mass limit for stochastic interacting particle systems with Lévy noise and linear alignment force.Chaos: An Interdisciplinary Journal of Nonlinear Science, 34(2), 2024

    work page 2024

  31. [31]

    J. Xu, J. Liu, J. Liu, and Y . Miao. Strong averaging principle for two-time-scale stochas- tic Mckean-Vlasov equations.Applied Mathematics & Optimization, 84(Suppl 1):837–867, 2021

    work page 2021

  32. [32]

    W. Xu, J. Duan, and W. Xu. An averaging principle for fractional stochastic differential equations with Lévy noise.Chaos: An Interdisciplinary Journal of Nonlinear Science, 30(8), 2020

    work page 2020

  33. [33]

    Zhao and W

    Q. Zhao and W. Wang. Convergence rate of Smoluchowski–Kramers approximation with stable Lévy noise.arXiv preprint arXiv:2411.11552, 2024. 20

    work page arXiv 2024