Probabilistic representation of solutions to the parabolic $p$-Laplace equation

Michael R\"ockner; Viorel Barbu

arxiv: 2604.26719 · v1 · submitted 2026-04-29 · 🧮 math.AP · math.PR

Probabilistic representation of solutions to the parabolic p-Laplace equation

Viorel Barbu , Michael R\"ockner This is my paper

Pith reviewed 2026-05-07 12:03 UTC · model grok-4.3

classification 🧮 math.AP math.PR

keywords equationlaplacenablasolutionsinftymathbbmathscrparabolic

0 comments

The pith

For p ≥ 4 and compactly supported initial data in L² with bounded gradient, the solution u to the parabolic p-Laplace equation is the law of a weak solution to a corresponding McKean-Vlasov SDE.

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

The p-Laplace equation models how a density spreads over time with a nonlinear speed that depends on the steepness of the density. The authors first establish a new regularity result showing the solution is smooth enough in a global sense. They then construct a stochastic process whose drift and diffusion coefficients depend on the current distribution of the process itself. The key result is that the probability distribution of this process at each time exactly matches the PDE solution.

Core claim

u(t,x) dx = ℒ_{X(t)}(dx), where X is a probabilistically weak solution to a McKean-Vlasov SDE, provided p ≥ 4, u₀ is a compactly supported probability density in L² with |∇u₀| ∈ L^∞.

Load-bearing premise

The new second-order global regularity result for weak solutions of the parabolic p-Laplace equation that is invoked to justify the stochastic representation and the well-posedness of the McKean-Vlasov equation.

read the original abstract

This work is concerned with the probabilistic representation of solutions to the $p$-Laplace evolution equation $\frac{\partial u}{\partial t}={\rm div}(|\nabla u|^{p-2}\nabla u)$ in $(0,\infty)\times\mathbb{R}^d$, $u(0,x)=u_0(x),$ $x\in\mathbb{R}^d$. One proves that, if $p\geq 4$, and if $u_0$ is a probability density with compact support and $u_0\in L^2$, $|\nabla u_0|\in L^\infty$, then $u$ can be represented as $u(t,x)dx=\mathscr L_{X(t)}(dx)$, where $\mathscr L_{X(t)}$ denotes the time marginal law of $X$ at time $t$ with $X$ being a probabilistically weak solution to a corresponding McKean-Vlasov stochastic differential equation. This result is based on a new second order global regularity result for the weak solutions to the parabolic $p$-Laplace equation.

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.

Desk Editor's Note private letter to a colleague

The paper gives a McKean-Vlasov SDE whose marginals recover solutions to the parabolic p-Laplace equation for p ≥ 4 under compactly supported initial data, but the argument stands or falls on a new global second-order regularity theorem.

read the letter

The core claim is that for p at least 4, with u0 a compactly supported probability density in L2 whose gradient is bounded, the PDE solution u(t,x) equals the law of a weak solution X to a McKean-Vlasov SDE whose coefficients are built from u itself. The marginals then match by construction once the SDE is well-posed. This is the probabilistic representation they establish, and it is new in this exact setting. The approach follows the usual pattern of defining the diffusion from the PDE solution rather than the other way around, so there is no circularity in the reduction. The conditions on p and u0 are chosen so the estimates close, and the abstract is explicit about what is assumed. That part of the work is clean and direct. The load-bearing step is the new second-order global regularity result for weak solutions. For the p-Laplacian with p > 2 the operator degenerates wherever the gradient vanishes, and uniform bounds on second derivatives over the whole space are not automatic. The stress-test note is right to flag that any gap in handling the degeneracy set or in upgrading local estimates to global ones would break both the SDE construction and the identification of marginals. Without the full proof it is impossible to check whether those estimates actually hold. This paper is aimed at people who work on stochastic representations of nonlinear parabolic equations or who want to explore particle approximations for degenerate diffusions. A reader already familiar with McKean-Vlasov techniques and the literature on p-Laplace regularity will get the most out of it. It deserves a serious referee because the representation itself is a concrete advance and the hypotheses are stated sharply, even if the regularity proof will need careful scrutiny.

Referee Report

2 major / 2 minor

Summary. The manuscript proves that, for p ≥ 4 and initial data u₀ that is a compactly supported probability density belonging to L² with |∇u₀| ∈ L^∞, the weak solution u of the parabolic p-Laplace equation admits the representation u(t,x) dx = ℒ_{X(t)}(dx), where X is a probabilistically weak solution of a McKean-Vlasov SDE whose coefficients are constructed from u. The argument rests on a new global second-order regularity theorem for weak solutions of the PDE.

Significance. If the regularity result holds, the work supplies a probabilistic representation for a class of degenerate parabolic equations, extending McKean-Vlasov techniques beyond the uniformly parabolic regime and potentially enabling particle approximations or Monte-Carlo schemes for the p-Laplace flow. The result is of interest to both PDE regularity theory and stochastic analysis of nonlinear diffusions.

major comments (2)

[Theorem on global regularity (Section 3)] The new global second-order regularity theorem (invoked throughout and stated as the sole new ingredient) is load-bearing: without uniform bounds on second derivatives that remain valid across the degeneracy set {∇u = 0}, the diffusion coefficient of the McKean-Vlasov SDE is not Lipschitz and the identification of marginal laws fails. The manuscript must supply the precise a-priori estimate (including the dependence on p and on the L^∞ bound of ∇u₀) that closes the global argument for p = 4.
[SDE construction and well-posedness (Section 4)] The construction of the McKean-Vlasov SDE (presumably in Section 4) defines its coefficients directly from the PDE solution u; the well-posedness proof therefore inherits every regularity gap. If the second-derivative bound is only local away from {∇u = 0}, the weak solution X may not exist globally or its marginals may not recover u.

minor comments (2)

[Introduction] The precise form of the McKean-Vlasov SDE (drift and diffusion coefficients) should be written explicitly already in the introduction, together with the Itô equation satisfied by X.
[Notation and preliminaries] Notation for the law ℒ_{X(t)} and for the probability space on which the weak solution lives should be introduced once and used consistently; several passages mix deterministic and stochastic notation without clear separation.

Axiom & Free-Parameter Ledger

0 free parameters · 2 axioms · 0 invented entities

The paper relies on standard existence theory for weak solutions of the p-Laplace equation and on well-posedness results for McKean-Vlasov SDEs; no new free parameters or invented entities are introduced in the abstract.

axioms (2)

domain assumption Existence of weak solutions to the parabolic p-Laplace equation under the given integrability conditions on u₀
Invoked implicitly to start the regularity analysis
domain assumption Well-posedness of the McKean-Vlasov SDE once the coefficients are defined from a sufficiently regular density
Required for the probabilistic representation to make sense

pith-pipeline@v0.9.0 · 5489 in / 1380 out tokens · 48443 ms · 2026-05-07T12:03:41.348698+00:00 · methodology

Probabilistic representation of solutions to the parabolic p-Laplace equation

Core claim

Load-bearing premise

discussion (0)