Pith. sign in

REVIEW 2 major objections 6 minor 1 cited by

Gradient estimates for nonlinear kinetic Fokker-Planck equations

T0 review · 2 major / 6 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 This paper proves that the velocity gradient of weak solutions to a broad class of nonlinear kinetic Fokker–Planck equations is controlled pointwise by the average gradient plus a truncated kinetic Riesz potential of the forcing term…

desk verdict Genuinely new gradient regularity results for nonlinear kinetic Fokker-Planck equations, but the comparison machinery rests on an unproved extension of a cited existence theorem; a serious referee should demand that gap be closed. read the letter →

arxiv 2502.09366 v1 pith:DOIBWZMY submitted 2025-02-13 math.AP

Kyeongbae Kim , Ho-Sik Lee , Simon Nowak This is my paper · ORCID
classification math.AP MSC 35K7035Q8435D3035B6531C45
keywords nonlinearkineticequationsFokker-PlanckultraparabolicSchauderestimatespotentialgradientregularityRieszpotentialsHöldercontinuity
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 a gradient regularity theory for nonlinear kinetic Fokker–Planck equations of the form ∂ₜf + v·∇ₓf − divᵥ(a(t,x,v,∇ᵥf)) = μ − divᵥG. Its central result is a pointwise bound: for Dini-continuous coefficients a and almost every point, |∇ᵥf(z₀)| is controlled by the average of |∇ᵥf| over a kinetic cylinder plus the truncated kinetic Riesz potential $I₁^{{|μ|}}$(z₀,R) of the nondivergence forcing μ. From this single estimate, the authors derive gradient continuity under a Lorentz-space condition, a VMO criterion, Calderón–Zygmund estimates, and Hölder regularity of ∇ᵥf under Hölder coefficients. The results are new already in the homogeneous case and even for linear divergence-form kinetic equations, and the homogeneous case partially answers an open problem. A sympathetic reader would care because it transfers the nonlinear potential-theoretic toolkit from elliptic and parabolic PDE to hypoelliptic kinetic equations.

What carries the argument

The load-bearing object is the truncated kinetic Riesz potential $I_α^{{|μ|}}$(z₀,R), which measures the singular mass of the forcing over kinetic cylinders, and the excess-decay estimate of Lemma 5.1, which shows that the velocity-gradient excess E(∇ᵥf; Q_{ρr}(z₀)) decays like ρ^α up to terms involving the coefficient modulus ω and the data mass. The mechanism that makes this work is the comparison of the given solution with solutions of the homogeneous equation with frozen coefficients (Lemmas 4.3–4.6), supported by a higher differentiability estimate in the spatial variable obtained through atomic-decomposition difference quotients. Together these produce the decay that is summed into a Riesz potential, converting the homogeneous gradient regularity into pointwise control for forced equations.

What would settle it

Verify Lemma 4.2 for a non-homogeneous flux such as a(ξ)=|ξ|^{p−2}ξ with p≠2 (which satisfies Assumption 1.1 but is not 1-homogeneous): if the initial-boundary value problem (4.3) fails to have a weak solution for some such a, then the comparison machinery and the pointwise estimates that rest on it collapse.

Watch

Extended reading notes

Core claim

On the paper's own terms, the central discovery is Theorem 1.9: if a is Dini-continuous and G≡0, then for almost every z₀ one has |∇ᵥf(z₀)| ≤ c(⨍_{Q_R(z₀)}|∇ᵥf| dz + $I₁^{{|μ|}}$(z₀,R)), where $I_α^{{|μ|}}$(z₀,R)=∫₀^R |μ|(Q_r(z₀))/$r^{{4n+2−α}}$ dr/r is the truncated kinetic Riesz potential. This pointwise bound is proved by comparing the solution with solutions of homogeneous constant-coefficient kinetic equations, establishing excess-decay estimates that convert the strong homogeneous decay into potential control of the gradient. The paper then shows that the bound yields, in a standard potential-theoretic way, gradient continuity when $I₁^{{|μ|}}$ vanishes at small scales (in particular when μ∈$L^{{4n+2,1}}$), VMO regularity under a weaker condition, Calderón–Zygmund estimates for q∈[2,4n+2), and, under Hölder-continuous coefficients, Hölder continuity of ∇ᵥf with exponent β∈(0,α) in the nonlinear case and any β∈(0,1) in the linear case. The homogeneous case is itself new: ∇ᵥf ∈ C^β_kin, which partially answers Problem 3 of [GN23].

Load-bearing premise

The paper's weakest load-bearing premise is Lemma 4.2, which asserts—citing only 'a careful inspection of the proof'—that the comparison problem (4.3) has a unique weak solution even after dropping the positive-homogeneity assumption on a; every comparison solution used in Sections 4 and 5 depends on that unproved extension.

Editorial extensions

If this is right

  • If the pointwise potential bound holds, then ∇ᵥf is continuous whenever the kinetic Riesz potential I₁^{|μ|} vanishes uniformly at small scales, in particular for μ ∈ L^{4n+2,1}.
  • Corollary 1.13 gives Calderón–Zygmund gradient estimates: ‖∇ᵥf‖_{L^{q(4n+2)/(4n+2−q)}} is controlled by the average gradient and the L^q norm of μ for every q∈[2,4n+2).
  • Under Hölder continuous coefficients, Theorem 1.14 yields M^#_{β,R}(∇ᵥf) bounds and the implication μ ∈ L^{(4n+2)/(1−β),∞} ⇒ ∇ᵥf ∈ C^β_kin; the linear version accepts any β∈(0,1).
  • For divergence data G, Theorem 1.16 shows BMO and C^β_kin regularity propagate from G to ∇ᵥf, and in the linear case f gains (1+β−ε)/3 Hölder regularity in x.
  • The homogeneous case (μ=G=0) itself yields ∇ᵥf ∈ C^β_kin for Hölder-continuous a, partially answering Problem 3 of [GN23].

Reading between the lines

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

  • A natural extension not taken in the paper is to admit measure-valued μ by replacing I₁^{|μ|} with a kinetic Wolff-type potential; the proof's use of |μ|(Q_r) masses suggests such an extension is plausible.
  • For the Landau equation, whose linearized operator is a kinetic Fokker–Planck operator in divergence form, these estimates suggest compactness and regularity criteria for the gradient in terms of L^{4n+2,1} control of the collision term.
  • The restriction β<α in the nonlinear homogeneous case likely reflects the De Giorgi iteration exponent rather than an intrinsic threshold; improving the comparison step might push the admissible β toward the linear range (0,1).
  • Because the proof constructs comparison solutions on regularized kinetic cylinders, the estimates should extend to boundary-value problems on Kolmogorov-type domains, where similar excess-decay arguments apply up to the boundary.
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

2 major / 6 minor

Summary. This paper develops a gradient regularity theory for nonlinear kinetic Fokker-Planck equations of the form ∂t f + v·∇x f − div_v(a(t,x,v,∇_v f)) = μ − div_v G, where a is uniformly elliptic in the gradient variable but otherwise rough. The main result is the pointwise bound on ∇_v f by the average of |∇_v f| plus a truncated kinetic Riesz potential of μ under Dini-continuous coefficients (Theorem 1.9), together with corollaries giving gradient continuity, VMO, Calderón-Zygmund and Hölder-type estimates (Corollaries 1.11–1.13, Theorems 1.14–1.17). The proof proceeds in three stages: Section 3 establishes Hölder regularity of ∇_x f and ∇_v f for homogeneous equations with coefficients independent of x,v; Section 4 derives zero- and first-order comparison estimates against frozen-coefficient problems; Section 5 iterates an excess decay to obtain the potential estimates. The homogeneous gradient Hölder theorem (Theorem 1.2) is also new and partially answers Problem 3 in [GN23].

Significance. If the proof is completed, these appear to be the first pointwise gradient estimates for nonlinear kinetic Fokker-Planck equations, and they extend the elliptic/parabolic nonlinear potential theory to a hypoelliptic kinetic setting. The homogeneous result is new even in the autonomous case, and the linear theorems complement the higher-order Schauder theory of Loher [Loh23]. The overall strategy is coherent, the comparison/decay structure is well organized, and the paper makes appropriate use of prior tools such as [GIMV19], [DKLN24a] and [DN25]; I see no circularity. The main weakness is a missing existence proof for the comparison solutions in Lemma 4.2, which is load-bearing for all subsequent comparison estimates; this gap is likely repairable but must be addressed before the central claims can be considered established.

major comments (2)
  1. [Section 4, Lemma 4.2] Lemma 4.2 is the sole existence result for the comparison problems (4.3) and (4.23), yet its proof consists of the assertion that 'a careful inspection' of [GN23, Theorem 1.5] shows existence after dropping condition (c), the positive homogeneity assumption a(t,x,v,λξ)=λa(t,x,v,ξ). The paper does not identify which steps of [GN23] require (c), and Assumption 1.1 alone is not obviously covered by that theorem. This is load-bearing: Lemmas 4.3, 4.5, 4.6, 5.1, 5.5 and 5.7 all use comparison solutions supplied by Lemma 4.2, so the excess-decay machinery and ultimately Theorems 1.9, 1.14 and 1.16 depend on this unproved extension. Please provide a self-contained existence and uniqueness proof under Assumption 1.1 (for example, a Galerkin or monotone-operator argument), or else restrict the main theorems to positively homogeneous nonlinearities.
  2. [Section 5.1, Lemma 5.1] The excess functional E(·;Q) used in (5.1) and repeatedly in the proof of Lemma 5.1 is never defined; Section 5.2 defines only the different functional E2. Since (5.1) is the core estimate from which Lemma 5.2 and Theorem 1.9 are derived, the statement is not checkable as written. Please define E explicitly and verify the properties used in (5.2)–(5.5), such as subadditivity, scaling behaviour, and the relation to the averages appearing in the subsequent decay arguments.
minor comments (6)
  1. [Section 1.1] 'This aim of this paper' should read 'The aim of this paper'.
  2. [Lemma 2.4] The displayed definition of z1 contains a typo: 'z1=(t1,x1,z2)' should presumably read 'z1=(t1,x1,v1)'.
  3. [Theorems 1.15 and 1.17] The phrase 'with respect β' should be 'with respect to β'; in Theorem 1.17 the cross-reference 'from Theorem 1.14' should be 'from Theorem 1.16', since the divergence-form estimates are in the latter theorem.
  4. [Theorem 1.16] The constant in the statement reads 'c = c(n, Λ, eβ, ωωω)'; the symbol 'eβ' appears to be a typo and should presumably be 'β'.
  5. [Lemmas 2.13 and 2.14] 'satsifies' is a typo for 'satisfies' in both lemmas.
  6. [Definition 1.8] The modulus of continuity is defined only for pairs with the same time variable; if this restriction is intentional, it would help to state this explicitly, since the name 'Dini continuous in W×V' might suggest a modulus controlling t as well.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the gradient estimates follow from a standard comparison argument, and the cited auxiliary results are independent and do not contain the target conclusions.

full rationale

The paper's derivation chain is not circular. Theorem 1.9 is obtained from comparison estimates in Section 5, which compare the solution with a homogeneous constant-coefficient problem whose gradient regularity is proved independently in Theorem 3.2 via difference-quotient and De Giorgi-type arguments. The data enter through the kinetic Riesz potential and maximal functions, and there is no fitted parameter that is later renamed as a prediction. The self-citations to [DKLN24a] and [DN25] are used only for auxiliary mapping properties of fractional maximal functions and Riesz potentials (Lemmas 2.7 and 2.8); those results have stated assumptions that do not include the paper's gradient estimates, and they are therefore independent support rather than load-bearing circularity. The only flagged concern is Lemma 4.2, whose proof states: 'a careful inspection of the proof reveals that the existence result given in [GN23, Theorem 1.5] remains true without the assumption (c).' All comparison solutions used in Sections 4 and 5 depend on this unproved extension. This is a genuine gap in completeness: if the extension fails, the excess-decay machinery collapses. However, it is not a circular step, because [GN23] is an external existence theorem that does not contain the present gradient conclusions, and the paper does not define any target quantity in terms of that theorem. Thus the derivation is self-contained against external benchmarks, and the circularity score is 0.

Assumptions & free parameters 0 free parameters · 7 assumptions · 0 invented entities

No free parameters are fitted to data; constants are structural functions of n, Λ, β, and the Dini modulus ω, not values tuned to force conclusions. The proofs rest on imported theorems from the kinetic regularity literature, listed above; none are introduced ad hoc to make the target estimate true. No new entities are postulated.

assumptions (7)
  • domain assumption Assumption 1.1: a is measurable in (t,x,v), C^1 in ξ, |a|≤Λ|ξ|, |∇_ξ a|≤Λ, and ∇_ξ a ζ·ζ ≥ Λ^{-1}|ζ|^2.
    Defines the class of nonlinearities studied; used throughout to linearize the equation via A=∫_0^1 ∇_ξ a(...,s∇_v f)ds.
  • standard math [GIMV19, Theorem 4] local boundedness and C^{α0} Hölder estimates for linear kinetic Fokker-Planck equations with rough coefficients.
    Basis of Lemma 2.11, which provides the zero-order regularity iterated on difference quotients in Section 3.
  • standard math [GN23, Theorem 1.5] existence and uniqueness of weak solutions for nonlinear kinetic initial-boundary value problems.
    Imported in Lemma 4.2 to generate comparison solutions w; the paper extends it by dropping homogeneity condition (c) without a full proof.
  • standard math [GIM24, Corollary 21] and [LN21] Poincaré inequality on kinetic cylinders; [GIMV19, Theorem 6] reverse Hölder inequality for ∇_v f.
    Used in Lemmas 2.13 and 2.14 to convert L1 control of ∇_v f into L2 and oscillation control.
  • standard math [IS20, Theorem 10.3] Lebesgue differentiation theorem for kinetic cylinders.
    Justifies the a.e. pointwise evaluation of the gradient in Theorem 1.9 and related results.
  • standard math [DKLN24a, Lemma 2.11] and [DN25, Proposition 2.5] mapping properties of kinetic Riesz potentials and fractional maximal functions.
    Used in Lemma 2.8 and Lemma 2.7 to convert pointwise estimates into Lorentz and Hölder corollaries.
  • standard math [FR24, Lemma A.1.2] interpolation for fractional difference quotients.
    Used in Lemma 3.1 and Theorem 3.2 to upgrade iterated Hölder estimates to C^1-type control.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Gradient estimates for nonlinear kinetic Fokker-Planck equations." pith.science (2026). https://pith.science/paper/DOIBWZMY

@misc{pith2026250209366,
  author       = {Pith},
  title        = {Pith review of: Gradient estimates for nonlinear kinetic Fokker-Planck equations},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/DOIBWZMY}},
  note         = {Machine review of arXiv:2502.09366}
}
read the original abstract

In this work, we provide a comprehensive gradient regularity theory for a broad class of nonlinear kinetic Fokker-Planck equations. We achieve this by establishing precise pointwise estimates in terms of the data in the spirit of nonlinear potential theory, leading to fine gradient regularity results under borderline assumptions on the data. Notably, our gradient estimates are novel already in the absence of forcing terms and even for linear kinetic Fokker-Planck equations in divergence form.

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 1 Pith paper

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. Kinetic Fokker-Planck equations with Maxwell boundary conditions

    math.AP 2026-07 conditional novelty 8.0 of 10

    For every α∈(0,1), solutions to the kinetic Fokker-Planck equation with Maxwell boundary conditions are C^{3/π·arccos(α/2)−1} up to the grazing set, and this exponent is optimal.

Reference graph

Works this paper leans on

14 extracted references · 9 canonical work pages · cited by 1 Pith paper

  1. [1]

    Weak and Perron Solutions for Stationary Kramers-Fokker-Planck Equations in Bounded Domains

    [AAMN24] D. Albritton, S. Armstrong, J. -C. Mourrat, and M. Novack. “Variational methods for the kinetic Fokker-Planck equation”. In: Anal. PDE 17.6 (2024), pp. 1953–2010. doi: 10.2140/apde.2024.17.1953. [AH24] B. Avelin and M. Hou. “Weak and Perron’s Solutions to Linear Kinetic Fokker- Planck Equations of Divergence Form in Bounded Domains”. In: arXiv:24...

    work page Pith review arXiv 2024

  2. [7]

    Boundedness of weak solutions to degenerate Kolmogorov equations of hypoelliptic type in bounded domains

    Courant Lecture Notes in Mathematics. Courant Institute of Mathematical Sciences, New York; American Mathematical Society, Providence, RI, 2011, pp. x+147. [H¨ or67] L. H¨ ormander. “Hypoelliptic second order differential equations”. In:Acta Math. 119 (1967), pp. 147–171. doi: 10.1007/BF02392081. [Hou24] M. Hou. “Boundedness of weak solutions to degenerat...

    work page Pith review arXiv 1967

  3. [10]

    Regularity of kinetic Fokker-Planck equations in bounded domains

    3934/dcds.2011.29.1261. [Zhu21] Y. Zhu. “Velocity averaging and H¨ older regularity for kinetic Fokker-Planck equa- tions with general transport operators and rough coefficients”. In: SIAM J. Math. Anal. 53.3 (2021), pp. 2746–2775. doi: 10.1137/20M1372147. [Zhu24] Y. Zhu. “Regularity of kinetic Fokker-Planck equations in bounded domains”. In: arXiv:2206.0...

    work page Pith review arXiv 2021

  4. [24]

    Kinetic maximal Lp-regularity with temporal weights and application to quasilinear kinetic diffusion equations

    doi: 10.1007/s00205-023-01884-7 . [NZ22] L. Niebel and R. Zacher. “Kinetic maximal Lp-regularity with temporal weights and application to quasilinear kinetic diffusion equations”. In: J. Differential Equations 307 (2022), pp. 29–82. doi: 10.1016/j.jde.2021.10.043. [PRS22] S. Polidoro, A. Rebucci, and B. Stroffolini. “Schauder type estimates for degenerate...

    work page doi:10.1007/s00205-023-01884-7 2022

  5. [31]

    American Mathematical Society, New York, 1948, pp

    American Mathematical Society Colloquium Publications. American Mathematical Society, New York, 1948, pp. xii+528. [HL11] Q. Han and F. Lin. Elliptic partial differential equations . Second. Vol

    work page 1948

  6. [37]

    REFERENCES 41 [Hil48] E

    doi: 10.3934/mine.2023043. REFERENCES 41 [Hil48] E. Hille. Functional Analysis and Semi-Groups. Vol. Vol

    doi:10.3934/mine.2023043

  7. [39]

    Potential theory for a class of strongly degenerate parabolic operators of Kolmogorov type with rough coefficients

    doi: 10.1016/j.jfa.2021.109226. [LN22] M. Litsg ˚ ard and K. Nystr¨ om. “Potential theory for a class of strongly degenerate parabolic operators of Kolmogorov type with rough coefficients”. In: J. Math. Pures Appl. (9) 157 (2022), pp. 45–100. doi: 10.1016/j.matpur.2021.11.004. [Loh23] A. Loher. “Quantitative Schauder estimates for hypoelliptic equations”....

    work page arXiv 2022

  8. [49]

    H¨ older estimates for kinetic Fokker-Planck equations up to the boundary

    Mathematical Surveys and Monographs. American Mathematical Society, Providence, RI, 1997, pp. xiv+278. doi: 10.1090/surv/049. [Sil22] L. Silvestre. “H¨ older estimates for kinetic Fokker-Planck equations up to the boundary”. In: Ars Inven. Anal. (2022), Paper No. 6,

    doi:10.1090/surv/049 2022

Show all 14 references
  1. [109]

    Wiley-VCH Verlag Berlin GmbH, Berlin, 2000, p

    Mathematical Research. Wiley-VCH Verlag Berlin GmbH, Berlin, 2000, p

    2000

  2. [245]

    Pointwise Calder´ on-Zygmund gradient estimates for the p-Laplace system

    doi: 10.1016/j.jde.2006.07.015. [BCDKS18] D. Breit, A. Cianchi, L. Diening, T. Kuusi, and S. Schwarzacher. “Pointwise Calder´ on-Zygmund gradient estimates for the p-Laplace system”. In: J. Math. Pures Appl. (9) 114 (2018), pp. 146–190. doi: 10.1016/j.matpur.2017.07.011. [BCM9...

    2018

  3. [256]

    A comparison estimate for singular p-Laplace equations and its consequences

    doi: 10.1002/3527600868. [NP23] Q.-H. Nguyen and N. C. Phuc. “A comparison estimate for singular p-Laplace equations and its consequences”. In: Arch. Ration. Mech. Anal. 247.3 (2023), Paper No. 49,

    2023 doi

  4. [361]

    Gradient Riesz potential estimates for a general class of measure data quasilinear systems

    [CKW23] I. Chlebicka, M. Kim, and M. Weidner. “Gradient Riesz potential estimates for a general class of measure data quasilinear systems”. In: arXiv:2307.15525 (2023). arXiv: 2307.15525 [math.AP]. [CM14] A. Cianchi and V. Maz’ya. “Gradient regularity via rearrangements for p-...

    2023 arXiv

  5. [2023]

    On the weak continuity of elliptic operators and applications to potential theory

    ©2023, pp. 101–148. doi: 10.1007/978-3-031-34089-5\_3 . [TW02] N. S. Trudinger and X. -J. Wang. “On the weak continuity of elliptic operators and applications to potential theory”. In: Amer. J. Math. 124.2 (2002), pp. 369–410. [Vil02] C. Villani. “A review of mathematical topi...

    2002 doi

  6. [2024]

    Gehring’s Lemma for kinetic Fokker-Planck equations

    [GIM24] J. Guerand, C. Imbert, and C. Mouhot. “Gehring’s Lemma for kinetic Fokker-Planck equations”. In: arXiv:2410.04933 (2024). arXiv: 2410.04933 [math.AP]. [GIMV19] F. Golse, C. Imbert, C. Mouhot, and A. F. Vasseur. “Harnack inequality for kinetic Fokker-Planck equations wi...

    2024 arXiv

Pith tools

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