Pith. sign in

REVIEW 1 major objections 4 minor 62 references

On the density of singular SDEs with fractional noise and applications to McKean-Vlasov equations

T0 review · 1 major / 4 minor · reviewed 2026-08-07 · deepseek-v4-flash

Pith's one-line read For SDEs driven by fractional Brownian motion with singular, distributional drifts, the solution's conditional law has a density with explicit Besov regularity and Gaussian tails, and these estimates prove new well-posedness results for…

desk verdict Strong Besov and McKean–Vlasov results, but the Gaussian-tail proof has a genuine uniformity gap in Lemma 6.6. read the letter →

arxiv 2506.11900 v1 pith:X74TZUYH submitted 2025-06-13 math.PR

classification math.PR MSC 60H5060H1060G2234A06
keywords StochasticdifferentialequationsfractionalBrownianmotionBesovregularityofdensitiesGaussiantailsregularizationbynoiseMcKean-Vlasovdistributionaldriftlocalnondeterminism
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 that fractional Brownian noise regularizes singular, possibly distributional drifts well enough for the solution's conditional law to have a density with quantitatively controlled Besov regularity: for $b\in L^q([0,T]; B^\gamma_\infty)$ satisfying $\gamma>1-1/(H(q'\lor 2))$, the conditional law $L(X_t|F_u)$ of the unique admissible solution (the limit of classical solutions for smooth approximating drifts) lies in $B^\eta_1$ for every $\eta<\gamma-1+1/(Hq')$, with $\|L(X_t|F_u)\|_{B^\eta_1}\leq C(1+(t-u)^{-\eta H})$ uniformly in the sample path. The same method yields space-time regularity of $t\mapsto L(X_t|F_u)$ in $L^{\tilde q}([u,T];B^\alpha_{1,1})$ for $\alpha<\min\{1/H+\gamma-1,1/(\tilde q H)\}$. For $H<1/2$ with $b\in L^\infty([0,T];B^\gamma_\infty)$ and $\gamma>1-1/(2H)$, the density obeys two-sided Gaussian bounds with the same scaling $t^{-Hd}\exp(-c t^{-2H}|x-x_0|^2)$ as pure fractional Brownian motion. These estimates are then used to prove existence of strong solutions to convolutional McKean-Vlasov equations for $\theta>1-1/H$, and pathwise uniqueness and uniqueness in law under additional Besov and initial-law conditions. A sympathetic reader should care because the paper supplies quantitative density and Gaussian-tail estimates for genuinely distributional drifts driven by fractional noise, and uses them to reach scaling-subcritical McKean-Vlasov thresholds.

What carries the argument

The load-bearing object is a conditional-law density criterion: for $X=\theta+G$, where $G$ is an $H$-locally nondeterministic Gaussian-Volterra process (so $\mathrm{Var}(G_t|F_s)\geq c_H(t-s)^{2H}$) and $\theta$ has conditional Hölder increments of order $\beta>H$, the conditional law $L(X_t|\mathcal G)$ belongs to $B^{\beta/H-1-}_1$ with norm growing only polynomially in $(t-u)^{-1}$. Two auxiliary mechanisms carry the quantitative statements: a shifted deterministic sewing lemma that bounds time-integrals of test functions against the conditional law, producing the space-time regularity via duality, and a local-nondeterminism estimate for Gaussian bridges, $\mathrm{Var}(P^{T,y}_t|F_\xi)\geq c_H(t-\xi)^{2H}(T-t)^{2H}/(T-\xi)^{2H}$, which prevents the bridge from becoming too deterministic near its endpoint. The a priori estimates on $\theta=X-B$ come from the stochastic sewing solution theory and are uniform in the time horizon and in the drift norm, which is what lets the scaling identity $X^\lambda_t=\lambda^{-H}X_{\lambda t}$ turn the unit-time estimate into the sharp $(t-u)^{-\eta H}$ singularity.

What would settle it

Set $b\equiv 0$; self-similarity predicts $\|L(B_t)\|_{B^\eta_1}\sim t^{-\eta H}$, which the paper's estimates reproduce exactly. To actually falsify, take a one-dimensional Riesz-type distributional drift $b(x)=|x|^{-s}$ with $H=0.4$ and $s$ just below the admissible threshold, approximate it smoothly, simulate the SDE, and check whether the $B^\eta_1$-norm of the conditional density grows strictly faster than $(t-u)^{-\eta H}$ for some $\eta<\gamma-1+1/(Hq')$; any growth rate above the claimed bound would refute Theorem 3.1.

Watch

Extended reading notes

Core claim

On the paper's own terms, the central discovery is that the decomposition $X=\theta+B$, with $\theta=X-B$ the drift part, reduces law regularity to two ingredients: conditional Hölder control of $\theta$ and local nondeterminism of the Gaussian driver. Under Assumption 1.1 the drift part obeys a priori conditional increment estimates, with constants independent of the time horizon and of the drift norm, and from this the conditional law's Besov regularity follows with the exponents stated in Theorem 3.1 and the time-space version in Theorem 4.1. For the Gaussian-tail result, the paper proves an $H$-local-nondeterminism estimate for fractional Brownian bridges, $\mathrm{Var}(P^{T,y}_t|F_\xi)\geq c_H(t-\xi)^{2H}(T-t)^{2H}/(T-\xi)^{2H}$, which is the missing input that lets the bridge representation of the density be controlled for distributional drifts; this yields the two-sided Gaussian bounds of Theorem 6.1. The McKean-Vlasov theorems follow by a bootstrap: the regularity of $\mu_t=L(Y_t)$ upgrades the convolution $b*\mu_t$, which then feeds back into the same Besov estimates, closing a local Grönwall argument.

Load-bearing premise

The load-bearing premise is that the solution theory the paper inherits, under the subcritical Besov condition on the drift, really provides pathwise uniqueness and a priori bounds on the drift part $X-B$ with constants independent of the time horizon and of the drift's norm; if those constants are not uniform, the scaling step that produces the sharp time exponents and the McKean-Vlasov bootstrap would break down.

Editorial extensions

If this is right

  • For any singular drift in the admissible class, the conditional law at time $t$ given $F_u$ is not just absolutely continuous but has explicitly measurable smoothness, with the same $(t-u)^{-\eta H}$ scaling as the pure fractional Brownian case.
  • The space-time Besov estimates put $t\mapsto L(X_t)$ in $L^1([0,T];B^{\alpha}_{1,1})$ for essentially every $\alpha<1/H$ when the drift is sufficiently regular, matching what fractional Brownian motion itself delivers; the paper records this as optimality in Remark 4.5.
  • For $H<1/2$, distributional drifts of Besov order $\gamma>1-1/(2H)$ give densities with two-sided Gaussian bounds that are sharp, so the solution's density is comparable to that of the driving fractional Brownian motion.
  • Convolutional McKean-Vlasov equations have strong solutions whenever $\theta>1-1/H$, a scaling-subcritical threshold, and the effective drift $b*\mu$ becomes Lipschitz, so the limiting particle system is classically well-posed.
  • Under the extra conditions of Theorem 1.7, among all solutions with $b*\mu\in L^1 C^1_b$, pathwise uniqueness and uniqueness in law hold, covering Coulomb and Riesz kernels in the stated Hurst regimes.

Reading between the lines

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

  • Editorial extension: the bridge local-nondeterminism bound is proven for general Gaussian-Volterra bridges, so the Gaussian-tail mechanism should transfer with minor changes to Riemann-Liouville or mixed fractional noises, giving two-sided densities for a wider family of singular Volterra SDEs.
  • Editorial extension: the McKean-Vlasov bootstrap uses only the time-integrated Besov regularity of the one-time marginals, so the same argument is likely to work for systems with several interaction kernels or for drift kernels depending on more than the current marginal law.
  • Editorial extension: the authors flag the initial-law condition $L(Y_0)\in L^\infty_x$ in the uniqueness theorem as possibly technical; a natural test is to see whether the propagation lemma can be run from $L^p$ initial densities, which would extend uniqueness to a larger class of initial data.
  • Editorial extension: the scaling heuristic in Remark 2.6 makes $\theta=1-1/H$ a plausible critical threshold for convolutional McKean-Vlasov equations; proving breakdown below this threshold for a simple kernel would confirm that the bootstrap is not just sufficient but necessary.
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

1 major / 4 minor

Summary. The paper studies the density of solutions to SDEs driven by fractional Brownian motion with singular, possibly distributional, drifts. It proves quantitative Besov regularity of conditional laws (Theorem 1.2, detailed as Theorems 3.1 and 4.1), uses these estimates to establish existence and uniqueness results for convolutional McKean–Vlasov equations (Theorems 1.6 and 1.7), and proves two-sided Gaussian bounds for the density when H∈(0,1/2) and the drift has Besov regularity γ>1−1/(2H) (Theorem 1.5, detailed as Theorem 6.1). The proofs proceed by abstract Gaussian-Volterra statements, deterministic and stochastic sewing arguments, scaling arguments, a bootstrap for the McKean–Vlasov equation, and a local nondeterminism analysis of Gaussian bridges.

Significance. If correct, the results are substantial. The Besov regularity estimates for conditional laws, with explicit time dependence, go beyond existing results for fractional noise; the McKean–Vlasov existence theorem lowers the required drift regularity to almost the scaling threshold; and the Gaussian bounds would be the first of their kind for genuinely distributional drifts in the fractional case. The paper is careful about uniform constants and about separating abstract Gaussian-Volterra mechanisms from fBm-specific estimates. However, the proof of the upper Gaussian bound contains a false uniformity claim in Lemma 6.6, so that part of the paper is not established as written. The Besov and McKean–Vlasov sections do not rely on this lemma.

major comments (1)
  1. [§6.2, Lemma 6.6 and its use in §6.3] The claimed uniform Fernique estimate (6.12) cannot hold. From (6.3), V^{T,y}=V_1−V_2 with V_1(t)=y K_H(T,t)/∫_0^T K_H(T,u)^2 du, so for any p<2, ||V_1||_{L^p[0,T]}=c_p|y| with c_p>0. The process V_2 is a centered Gaussian process and hence symmetric. Writing X=V_1+V_2, symmetry of V_2 gives E[exp(c||X||^2)] = E[exp(c||V_1−V_2||^2)], and convexity of u↦exp(cu^2) yields E[exp(c||V_1+V_2||^2)] ≥ exp(c||V_1||^2), which is unbounded as |y|→∞. Therefore no positive constant c_p can work uniformly in y. This invalidates the use of Lemma 6.6 in Proposition 6.10, specifically the bound (6.23), and consequently the step in the proof of Theorem 6.1 stating that Proposition 6.10 readily implies sup_{y∈R^d} Ψ(T,y)<∞. The upper Gaussian bound in Theorem 6.1 is therefore not proved as written. The lower bound and the Besov/McKean–Vlasov results do not depend on Lemma 6.6. A possible repair is to replace Lemma 6.6 by an estimate with explicit y-dependence, such as E[exp(c||V^{T,y}||^2_{L^p})] ≤ C exp(C'|y|^2), and then choose the constants in Proposition 6.10 so that the extra factor is absorbed by the Gaussian factor exp(−|y|^2/(2T^{2H})) in (6.8); this needs to be carried out explicitly.
minor comments (4)
  1. [§6.3, Proposition 6.5, Eq. (6.8)] The prefactor in the displayed formula for p(T,y) is written as (2πT)^{−dH}; the correct Gaussian normalization for a d-dimensional fBm with variance T^{2H} is (2π)^{−d/2} T^{−dH} or equivalently (2π T^{2H})^{−d/2}. This does not affect the qualitative Gaussian bounds because the constants in Theorem 6.1 can absorb the error, but the exact formula should be corrected.
  2. [§5.1, proof of Theorem 5.1] The proof chooses parameters with γ:=θ+α>1 immediately before invoking Lemma 5.4. The proof of Lemma 5.4, however, applies Theorem 4.1 only when the drift regularity γ is <1; the a priori estimate should be applied with a smaller δ (so that γ<1), and the passage to γ>1 should happen only when identifying the limit b∗μ∞∈L^ρ([0,T];B^γ_∞). The statement of Lemma 5.4 covers the needed range, so this is a clarity issue, but the current text is misleading.
  3. [§5.1, Lemma 5.4] In the proof of Lemma 5.4, the sentence 'γ fulfills Assumption 1.1 for q=ρ' is only true if δ is chosen smaller than 1/(ρ′H); otherwise γ may be ≥1. The proof should state explicitly that δ is chosen small enough so that 1−1/(ρ′H)+δ<1.
  4. [§6.3, proof of Theorem 6.1] The lower-bound argument uses Jensen's inequality and then bounds E[||L^{T,y}||^2_{L^{2+ε}}] using Lemma 6.7. For this step it is important that the exponential estimate in Lemma 6.7 is uniform in y, since otherwise the bound would depend on y in an uncontrolled way. The text should state this uniformity explicitly when deriving the lower bound.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the central Besov and Gaussian estimates are derived from stated assumptions and independently proven a priori bounds; the McKean-Vlasov bootstrap is a closed Grönwall argument; the imported well-posedness from [30] is an external theorem, not an assumption of the target results.

full rationale

I traced the derivation chain for Theorem 1.2(a)-(b) (Theorems 3.1 and 4.1), the McKean-Vlasov theorems (5.1, 5.2), and the Gaussian bounds (6.1). The Besov regularity estimates are proven from Assumption 1.1, the local nondeterminism of Gaussian-Volterra processes, Romito's lemma extended to conditional laws (Lemma 3.5), and a priori estimates for φ = X − B that the paper proves in Lemmas 2.8 and 2.11 as refinements of [30]; they are not assumed in the target bound. The (t−u)^{−ηH} singularity comes from the scaling identity (3.12) and the Besov scaling Lemma A.2, not from a fitted exponent. The space-time estimates use a shifted deterministic sewing lemma (Lemma 2.7) proved in Appendix C, and the duality argument does not presuppose the conclusion. In Section 5, the McKean-Vlasov a priori estimate (Lemma 5.4) is a bootstrap: the unknown norm appears on both sides, but the small-time coefficients allow closure via the local Grönwall inequality (5.7)-(5.9). This is a standard self-improvement, not an identity, and the compactness step then yields existence. The invocation of [30] for well-posedness of the linear SDE and for stability estimates is load-bearing, but it is an external published theorem with assumptions that do not include the paper's target results; although an author overlaps, the citation is genuine evidence. The Gaussian-bounds section relies on exponential estimates for the bridge drift; the skeptic's objection to Lemma 6.6 is a potential mathematical error in the uniformity claim, but it is not a circularity because the target bound is not assumed in proving Lemma 6.6 and no equation in Section 6 reduces to itself by construction. I found no self-definitional reduction, no fitted-parameter-as-prediction, and no renaming of a known result.

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

No free parameters are fitted: the paper introduces no constants tuned to reproduce the target estimates. All bounds are a priori with explicit dependence on norms of b. No new physical or mathematical entities are postulated; the shifted sewing lemma (Lemma 2.7) and bridge LND (Proposition 6.3) are new lemmas about existing mathematical objects.

assumptions (4)
  • domain assumption Existence, uniqueness, and stability of admissible solutions to SDE(b,X_0) under Assumption 1.1, imported from Galeati-Gerencsér [30, Theorem 5.6].
    Invoked in Remark 2.3 and used throughout; the entire paper operates in the well-posedness regime established by [30].
  • standard math Fractional Brownian motion admits the Volterra representation B_t = ∫_0^t K_H(t,s) dW_s with an invertible kernel on its Cameron-Martin space (Nualart-Ouknine [49]).
    Used for the Girsanov transform in Proposition 6.5 and for the definition of K_H^{-1}; the explicit kernel for H ∈ (0,1/2) is given in (6.7).
  • standard math Local nondeterminism of fBm: Var(B_t|F_s) = C_H (t-s)^{2H} (equation 2.3), and the analogous lower bound for Gaussian-Volterra processes defined in (2.7).
    This is the quantitative backbone for all density estimates; it appears in Sections 3, 4, and 6, and in the new bridge LND estimate Corollary 6.4.
  • standard math Besov embedding, duality, convolution, and heat-kernel estimates collected in Appendix A (Lemmas A.1-A.9).
    These classical results are used pervasively, e.g. for Lemma A.9(c) in the shifted sewing argument and for Corollary A.8 in the uniqueness proof.

how reviews work

0 comments
Cite this review

Pith. "Pith review of On the density of singular SDEs with fractional noise and applications to McKean-Vlasov equations." pith.science (2026). https://pith.science/paper/X74TZUYH

@misc{pith2026250611900,
  author       = {Pith},
  title        = {Pith review of: On the density of singular SDEs with fractional noise and applications to McKean-Vlasov equations},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/X74TZUYH}},
  note         = {Machine review of arXiv:2506.11900}
}
read the original abstract

We investigate properties of the (conditional) law of the solution to SDEs driven by fractional Brownian noise with a singular, possibly distributional, drift. Our results on the law are twofold: i) we quantify the spatial regularity of the law, while keeping track of integrability in time, and ii) we prove that it has a density with Gaussian tails. Then the former result is used to establish novel results on existence and uniqueness of solutions to McKean-Vlasov equations of convolutional type.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

62 extracted references · 55 canonical work pages

  1. [1]

    thesis, Universit´ e Paris-Saclay, 2023, https://theses.hal.science/ tel-04576910/

    Lukas Anzeletti, Stochastic differential equations with singular drift and (fractional) Brow- nian noise , Ph.D. thesis, Universit´ e Paris-Saclay, 2023, https://theses.hal.science/ tel-04576910/

  2. [2]

    Lukas Anzeletti, Alexandre Richard, and Etienne Tanr´ e, Regularisation by fractional noise for one-dimensional differential equations with distributional drift , Electron. J. Probab. 28 (2023), Paper No. 135, 49, doi. MR-4664455

  3. [3]

    Pure Appl

    Siva Athreya, Oleg Butkovsky, Khoa Lˆ e, and Leonid Mytnik, Well-posedness of stochastic heat equation with distributional drift and skew stochastic heat equation , Comm. Pure Appl. Math. 77 (2024), no. 5, 2708–2777, doi. MR-4720220

  4. [4]

    343, Springer, Heidelberg, 2011, doi

    Hajer Bahouri, Jean-Yves Chemin, and Rapha¨ el Danchin, Fourier analysis and nonlinear partial differential equations, Grundlehren der Mathematischen Wissenschaften [Fundamental Principles of Mathematical Sciences], vol. 343, Springer, Heidelberg, 2011, doi. MR-2768550

  5. [5]

    Isma¨ el Bailleul, R´ emi Catellier, and Fran¸ cois Delarue,Solving mean field rough differential equations, Electron. J. Probab. 25 (2020), Paper No. 21, 51, doi. MR-4073682 45

  6. [6]

    , Propagation of chaos for mean field rough differential equations , Ann. Probab. 49 (2021), no. 2, 944–996, doi. MR-4255135

  7. [7]

    Related Fields 13 (2007), no

    Fabrice Baudoin and Laure Coutin, Volterra bridges and applications, Markov Process. Related Fields 13 (2007), no. 3, 587–596. MR-2357390

  8. [8]

    Fabrice Baudoin, Eulalia Nualart, Cheng Ouyang, and Samy Tindel, On probability laws of solutions to differential systems driven by a fractional Brownian motion , Ann. Probab. 44 (2016), no. 4, 2554–2590, doi. MR-3531675

Show all 62 references
  1. [9]

    Mireia Besal´ u, Arturo Kohatsu-Higa, and Samy Tindel, Gaussian-type lower bounds for the density of solutions of SDEs driven by fractional Brownian motions , Ann. Probab. 44 (2016), no. 1, 399–443, doi. MR-3456342

  2. [10]

    Adrien Blanchet, Jean Dolbeault, and Beno ˆ ıt Perthame,Two-dimensional Keller-Segel model: optimal critical mass and qualitative properties of the solutions , Electron. J. Differential Equa- tions (2006), no. 44, 1–33. MR-2226917

  3. [11]

    Bogachev, Gaussian measures, Mathematical Surveys and Monographs, vol

    Vladimir I. Bogachev, Gaussian measures, Mathematical Surveys and Monographs, vol. 62, American Mathematical Society, Providence, RI, 1998, doi. MR-1642391

  4. [12]

    Maximilian Buthenhoff and Ercan S¨ onmez, Gaussian-type density estimates for mixed SDEs driven by correlated fractional Brownian motions , 2025, https://arxiv.org/abs/2503. 03685

  5. [13]

    Oleg Butkovsky and Samuel Gallay, Weak existence for SDEs with singular drifts and frac- tional Brownian or L´ evy noise beyond the subcritical regime, 2023, https://arxiv.org/abs/ 2311.12013

  6. [14]

    Oleg Butkovsky, Khoa Lˆ e, and Leonid Mytnik,Stochastic equations with singular drift driven by fractional Brownian motion, To appear in Prob. Math. Phys. (2023), https://arxiv.org/ abs/2302.11937

  7. [15]

    Oleg Butkovsky and Leonid Mytnik, Weak uniqueness for singular stochastic equations , 2024, https://arxiv.org/abs/2405.13780

  8. [16]

    Thomas Cass and Peter Friz, Densities for rough differential equations under H¨ ormander’s condition, Ann. of Math. (2) 171 (2010), no. 3, 2115–2141, doi. MR-2680405

  9. [17]

    R´ emi Catellier and Romain Duboscq,Regularization by noise for rough differential equations driven by Gaussian rough paths , Ann. Probab. 53 (2025), no. 1, 79–139, doi. MR-4852889

  10. [18]

    R´ emi Catellier and Massimiliano Gubinelli,Averaging along irregular curves and regularisation of ODEs , Stochastic Process. Appl. 126 (2016), no. 8, 2323–2366, doi. MR-3505229

  11. [19]

    Paul-Eric Chaudru de Raynal, Strong well posedness of McKean-Vlasov stochastic differential equations with H¨ older drift, Stochastic Process. Appl. 130 (2020), no. 1, 79–107, doi. MR- 4035024

  12. [20]

    Partial Differ

    Paul-Eric Chaudru de Raynal, Jean-Fran¸ cois Jabir, and St´ ephane Menozzi,Multidimensional stable driven McKean–Vlasov SDEs with distributional interaction kernel: a regularization by noise perspective, Stoch. Partial Differ. Equ. Anal. Comput. 13 (2025), no. 1, 367–420, doi....

  13. [21]

    Friz, and Mario Maurelli, Pathwise McKean-Vlasov theory with additive noise , Ann

    Michele Coghi, Jean-Dominique Deuschel, Peter K. Friz, and Mario Maurelli, Pathwise McKean-Vlasov theory with additive noise , Ann. Appl. Probab. 30 (2020), no. 5, 2355–2392, doi. MR-4149531

  14. [22]

    Reine Angew

    Gianluca Crippa and Camillo De Lellis, Estimates and regularity results for the DiPerna-Lions flow, J. Reine Angew. Math. 616 (2008), 15–46, doi. MR-2369485 46

  15. [23]

    Konstantinos Dareiotis and M´ at´ e Gerencs´ er,Path-by-path regularisation through multiplicative noise in rough, Young, and ordinary differential equations , Ann. Probab. 52 (2024), no. 5, 1864–1902, doi. MR-4791421

  16. [24]

    Konstantinos Dareiotis, M´ at´ e Gerencs´ er, Khoa Lˆ e, and Chengcheng Ling,Regularisation by Gaussian rough path lifts of fractional Brownian motions , 2024, https://arxiv.org/abs/ 2412.01645

  17. [25]

    Laurent Decreusefond, Stochastic integration with respect to Volterra processes, Ann. Inst. H. Poincar´ e Probab. Statist.41 (2005), no. 2, 123–149, doi. MR-2124078

  18. [26]

    Eleonora Di Nezza, Giampiero Palatucci, and Enrico Valdinoci, Hitchhiker’s guide to the fractional Sobolev spaces, Bull. Sci. Math. 136 (2012), no. 5, 521–573, doi. MR-2944369

  19. [27]

    [Saint-Flour Probability Summer School], Lecture Notes in Mathematics, vol

    Franco Flandoli, Random perturbation of PDEs and fluid dynamic models , Lectures from the 40th Probability Summer School held in Saint-Flour, 2010, ´Ecole d’ ´Et´ e de Probabilit´ es de Saint-Flour. [Saint-Flour Probability Summer School], Lecture Notes in Mathematics, vol. 20...

  20. [28]

    Nicolas Fournier and Benjamin Jourdain, Stochastic particle approximation of the Keller- Segel equation and two-dimensional generalization of Bessel processes , Ann. Appl. Probab. 27 (2017), no. 5, 2807–2861, doi. MR-3719947

  21. [29]

    Friz and Nicolas B

    Peter K. Friz and Nicolas B. Victoir, Multidimensional stochastic processes as rough paths , Cambridge Studies in Advanced Mathematics, vol. 120, Cambridge University Press, Cam- bridge, 2010, doi. MR-2604669

  22. [30]

    Sigma 13 (2025), Paper No

    Lucio Galeati and M´ at´ e Gerencs´ er,Solution theory of fractional SDEs in complete subcritical regimes, Forum Math. Sigma 13 (2025), Paper No. e12, doi. MR-4854429

  23. [31]

    Harang, and Avi Mayorcas, Distribution dependent SDEs driven by additive continuous noise, Electron

    Lucio Galeati, Fabian A. Harang, and Avi Mayorcas, Distribution dependent SDEs driven by additive continuous noise, Electron. J. Probab. 27 (2022), Paper No. 37, 38, doi. MR-4388460

  24. [32]

    Theory Related Fields 185 (2023), no

    , Distribution dependent SDEs driven by additive fractional Brownian motion , Probab. Theory Related Fields 185 (2023), no. 1-2, 251–309, doi. MR-4528970

  25. [33]

    org/abs/2403.05454

    Lucio Galeati, Khoa Lˆ e, and Avi Mayorcas, Quantitative Propagation of Chaos for Singular Interacting Particle Systems Driven by Fractional Brownian Motion , 2024, https://arxiv. org/abs/2403.05454

  26. [34]

    2, Springer, Berlin, 2007, pp

    Dario Gasbarra, Tommi Sottinen, and Esko Valkeila, Gaussian Bridges , Stochastic analysis and applications, Abel Symp., vol. 2, Springer, Berlin, 2007, pp. 361–382, doi. MR-2397795

  27. [35]

    Partial Differ

    M´ at´ e Gerencs´ er,Regularisation by regular noise , Stoch. Partial Differ. Equ. Anal. Comput. 11 (2023), no. 2, 714–729, doi. MR-4588621

  28. [36]

    Pillai, Regularity of laws and ergodicity of hypoelliptic SDEs driven by rough paths , Ann

    Martin Hairer and Natesh S. Pillai, Regularity of laws and ergodicity of hypoelliptic SDEs driven by rough paths , Ann. Probab. 41 (2013), no. 4, 2544–2598, doi. MR-3112925

  29. [37]

    Yi Han, Smoothness of the density for McKean-Vlasov SDEs with measurable kernel , 2022, https://arxiv.org/abs/2208.02771

  30. [38]

    Yi Han, Solving McKean–Vlasov SDEs via relative entropy , Ann. Appl. Probab. 35 (2025), no. 2, 858–897, doi. MR-4897749

  31. [39]

    Zimo Hao, Michael R¨ ockner, and Xicheng Zhang,Distribution-flow dependent SDEs driven by (fractional) Brownian motion and Navier-Stokes equations , 2024, https://arxiv.org/abs/ 2405.19034

  32. [40]

    Zimo Hao and Xicheng Zhang, SDEs with supercritical distributional drifts , 2024, https: //arxiv.org/abs/2312.11145. 47

  33. [41]

    Tuomas Hyt¨ onen, Jan van Neerven, Mark Veraar, and Lutz Weis,Analysis in Banach spaces. Vol. I. Martingales and Littlewood-Paley theory , Ergebnisse der Mathematik und ihrer Gren- zgebiete. 3. Folge. A Series of Modern Surveys in Mathematics [Results in Mathematics and Relate...

  34. [42]

    Schilling,Convolution inequalities for Besov and Triebel-Lizorkin spaces, and applications to convolution semigroups , Studia Math

    Franziska K¨ uhn and Ren´ e L. Schilling,Convolution inequalities for Besov and Triebel-Lizorkin spaces, and applications to convolution semigroups , Studia Math. 262 (2022), no. 1, 93–119, doi. MR-4339468

  35. [43]

    Khoa Lˆ e,A stochastic sewing lemma and applications , Electron. J. Probab. 25 (2020), Paper No. 38, 55, doi. MR-4089788

  36. [44]

    181, American Mathematical Society, Providence, RI, 2017, doi

    Giovanni Leoni, A first course in Sobolev spaces, second ed., Graduate Studies in Mathematics, vol. 181, American Mathematical Society, Providence, RI, 2017, doi. MR-3726909

  37. [45]

    Xue-Mei Li, Fabien Panloup, and Julian Sieber, On the (non)stationary density of fractional- driven stochastic differential equations , Ann. Probab. 51 (2023), no. 6, 2056–2085, doi. MR- 4666290

  38. [46]

    Toyomu Matsuda and Avi Mayorcas,Pathwise Uniqueness for Multiplicative Young and Rough Differential Equations Driven by Fractional Brownian Motion , 2023, https://arxiv.org/ abs/2312.06473

  39. [47]

    Sigma 12 (2024), Paper No

    Toyomu Matsuda and Nicolas Perkowski, An extension of the stochastic sewing lemma and applications to fractional stochastic calculus , Forum Math. Sigma 12 (2024), Paper No. e52, 52, doi. MR-4730255

  40. [48]

    Viens, Density formula and concentration inequalities with Malli- avin calculus, Electron

    Ivan Nourdin and Frederi G. Viens, Density formula and concentration inequalities with Malli- avin calculus, Electron. J. Probab. 14 (2009), no. 78, 2287–2309, doi. MR-2556018

  41. [49]

    David Nualart and Youssef Ouknine, Regularization of differential equations by fractional noise, Stochastic Process. Appl. 102 (2002), no. 1, 103–116, doi. MR-1934157

  42. [50]

    Tudor, Existence and Besov regularity of the density for a class of SDEs with Volterra noise , C

    Christian Olivera and Ciprian A. Tudor, Existence and Besov regularity of the density for a class of SDEs with Volterra noise , C. R. Math. Acad. Sci. Paris 357 (2019), no. 7, 636–645, doi. MR-3998291

  43. [51]

    Andrea Pascucci and Alessio Rondelli, McKean-Vlasov stochastic equations with H¨ older coef- ficients, Stochastic Process. Appl. 182 (2025), Paper No. 104564, 5, doi. MR-4848654

  44. [52]

    Andrea Pascucci, Alessio Rondelli, and Alexander Yu Veretennikov, Existence and uniqueness results for strongly degenerate McKean-Vlasov equations with rough coefficients, 2024, https: //arxiv.org/abs/2409.14451

  45. [53]

    59 (2023), no

    Nicolas Perkowski and Willem van Zuijlen, Quantitative heat-kernel estimates for diffusions with distributional drift , Potential Anal. 59 (2023), no. 2, 731–752, doi. MR-4625000

  46. [54]

    Pitt, Local Times for Gaussian Vector Fields , Indiana Univ

    Loren D. Pitt, Local Times for Gaussian Vector Fields , Indiana Univ. Math. J. 27 (1978), no. 2, 309–330, doi. MR-471055

  47. [55]

    2, 1131–1158, doi

    Michael R¨ ockner and Xicheng Zhang, Well-posedness of distribution dependent SDEs with singular drifts , Bernoulli 27 (2021), no. 2, 1131–1158, doi. MR-4255229

  48. [56]

    Marco Romito, A simple method for the existence of a density for stochastic evolutions with rough coefficients, Electron. J. Probab. 23 (2018), Paper no. 113, 43, doi. MR-3885546

  49. [57]

    Samko, Anatoly A

    Stefan G. Samko, Anatoly A. Kilbas, and Oleg I. Marichev, Fractional integrals and deriva- tives, Theory and applications, Gordon and Breach Science Publishers, Yverdon, 1993. MR- 1347689 48

  50. [58]

    1464, Springer, Berlin, 1991, pp

    Alain-Sol Sznitman, Topics in propagation of chaos , ´Ecole d’ ´Et´ e de Probabilit´ es de Saint- Flour XIX—1989, Lecture Notes in Math., vol. 1464, Springer, Berlin, 1991, pp. 165–251, doi. MR-1108185

  51. [59]

    78, Birkh¨ auser Verlag, Basel, 1983, doi

    Hans Triebel, Theory of function spaces , Monographs in Mathematics, vol. 78, Birkh¨ auser Verlag, Basel, 1983, doi. MR-781540

  52. [60]

    Alexander Veretennikov, On weak existence of solutions of degenerate McKean-Vlasov equa- tions, Stoch. Dyn. 24 (2024), no. 5, Paper No. 2450032, 15, doi. MR-4832751

  53. [61]

    338, Springer-Verlag, Berlin, 2009, doi

    C´ edric Villani, Optimal transport , Old and new, Grundlehren der mathematischen Wis- senschaften [Fundamental Principles of Mathematical Sciences], vol. 338, Springer-Verlag, Berlin, 2009, doi. MR-2459454

  54. [62]

    Xicheng Zhang and Guohuan Zhao, Heat kernel and ergodicity of SDEs with distributional drifts, 2018, https://arxiv.org/abs/1710.10537. 49

Pith tools

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