Jump to ContentJump to Main Navigation
Show Summary Details
More options …

Advances in Nonlinear Analysis

Editor-in-Chief: Radulescu, Vicentiu / Squassina, Marco

IMPACT FACTOR 2018: 6.636

CiteScore 2018: 5.03

SCImago Journal Rank (SJR) 2018: 3.215
Source Normalized Impact per Paper (SNIP) 2018: 3.225

Mathematical Citation Quotient (MCQ) 2018: 3.18

Open Access
See all formats and pricing
More options …

Regularity of solutions of the parabolic normalized p-Laplace equation

Fredrik Arbo Høeg
  • Corresponding author
  • Department of Mathematical Sciences, Norwegian University of Science and Technology, NO-7491 Trondheim, Norway
  • Email
  • Other articles by this author:
  • De Gruyter OnlineGoogle Scholar
/ Peter Lindqvist
Published Online: 2018-07-12 | DOI: https://doi.org/10.1515/anona-2018-0091


The parabolic normalized p-Laplace equation is studied. We prove that a viscosity solution has a time derivative in the sense of Sobolev belonging locally to L2.

Keywords: Non-linear equation; regularity theory; time derivative

MSC 2010: 35K92; 35K10

1 Introduction

We consider viscosity solutions of the normalized p-Laplace equation


in ΩT=Ω×(0,T), Ω being a domain in n. Formally, the equation reads


In the linear case p=2, we have the heat equation ut=Δu, and also for n=1, the equation reduces to the heat equation ut=(p-1)uxx. At the limit p=1, we obtain the equation for motion by mean curvature. We aim at showing that the time derivative ut exists in the Sobolev sense and belongs to Lloc2(ΩT). We also study the second derivatives 2uxixj.

There has been some recent interest in connection with stochastic game theory, where the equation appears, cf. [7]. From our point of view, the work [3] is of actual interest, because there it is shown that the time derivative ut of the viscosity solutions exists and is locally bounded, provided that the lateral boundary values are smooth. Thus, the boundary values control the time regularity. If no such assumptions about the behaviour at the lateral boundary Ω×(0,T) are made, a conclusion like utLloc(ΩT) is in doubt. Our main result is the following, where we unfortunately have to restrict p.

Theorem 1.1.

Suppose that u=u(x,t) is a viscosity solution of the normalized p-Laplace equation in ΩT. If 65<p<145, then the Sobolev derivatives ut and 2uxixj exist and belong to Lloc2(ΩT).

We emphasize that no assumptions on the boundary values are made for this interior estimate. Our method of proof is based on a verification of the identity


where we have to prove that the function U, which is the right-hand side of equation (1.1), belongs to Lloc2(ΩT). Thus, the second spatial derivatives D2u are crucial (local boundedness of u was proven in [3, 2] and interior Hölder estimates for the gradient in [6]). The elliptic case has been studied in [1].

In the range 1<p<2, one can bypass the question of second derivatives.

Theorem 1.2.

Suppose that u=u(x,t) is a viscosity solution of the normalized p-Laplace equation in ΩT. If 1<p<2, then the Sobolev derivative ut exists and belongs to Lloc2(ΩT).

To avoid the problem of vanishing gradient, we first study the regularized equation


Here the classical parabolic regularity theory is applicable. The equation was studied by Does in [3], where an estimate of the gradient uϵ was found with Bernstein’s method. We shall prove a maximum principle for the gradient. Further, we differentiate equation (1.2) with respect to the space variables and derive estimates for uϵ, which are passed over to the solution u of (1.1).

Analogous results seem to be possible to reach through the Cordes condition. This also restricts the range of valid exponents p. We have refrained from this approach, mainly since the absence of zero (lateral) boundary values produces many undesired terms to estimate. Finally, we mention that the limits 65 and 145 in Theorem 1.1 are evidently an artifact of the method. It would be interesting to know whether the theorem is valid in the whole range 1<p<. In any case, our method is not capable to reach all exponents.

2 Preliminaries


The gradient of a function f:ΩT is


and its Hessian matrix is


We shall, occasionally, use the abbreviation


for partial derivatives. Young’s inequality


is often referred to. Finally, the summation convention is used when convenient.

Viscosity solutions.

The normalized p-Laplace equation is not in divergence form. Thus, the concept of weak solutions with test functions under the integral sign is problematic. Fortunately, the modern concept of viscosity solutions works well. The existence and uniqueness of viscosity solutions of the normalized p-Laplace equation was established in [2]. We recall the definition.

Definition 2.1.

We say that an upper semi-continuous function u is a viscosity subsolution of equation (1.1) if for all ϕC2(ΩT), we have


at any interior point (x,t) where u-ϕ attains a local maximum, provided ϕ(x,t)0. Further, at any interior point (x,t) where u-ϕ attains a local maximum and ϕ(x,t)=0, we require


for some ηn, with |η|1.

Definition 2.2.

We say that a lower semi-continuous function u is a viscosity supersolution of equation (1.1) if for all ϕC2(ΩT), we have


at any interior point (x,t) where u-ϕ attains a local minimum, provided ϕ(x,t)0. Further, at any interior point (x,t) where u-ϕ attains a local minimum and ϕ(x,t)=0, we require


for some ηn, with |η|1.

Definition 2.3.

A continuous function u is a viscosity solution if it is both a viscosity subsolution and a viscosity supersolution.

For a detailed discussion on the definition at critical points, we refer to [5]. The reason behind the choice of ηn is given in [5, Section 2]. The viscosity solutions of equation (1.2) are defined in a similar manner, except that now ϕ(x,t)=0 is not a problem.

Maximum principle for the gradient.

In order to estimate the time derivative, we need bounds on the second derivatives of uϵ (and also on its gradient). If we first assume that uϵ is C1 on the parabolic boundary parΩT, we get bounds on the gradient in all of ΩT. This follows from the following maximum principle.

Proposition 2.4 (Maximum principle).

Let uϵ be a solution of equation (1.2). If uϵC1(Ω¯T), then



With some modifications, a proof can be extracted from [3]. We give a direct proof. To this end, consider


To find the partial differential equation satisfied by Vϵ, we calculate1


Writing equation (1.1) in the form


we find


Rearranging and using


we arrive at the following differential equation for Vϵ:



w(x,t)=|uϵ(x,t)|2+ϵ2-αt=Vϵ(x,t)-αtfor α>0.

Suppose that wϵ has an interior maximum point at (x0,t0). At this point, Vϵ(x0,t0)>0, otherwise we would have Vϵ(x,t)0 in ΩT, in which case there is nothing to prove. By the infinitesimal calculus,


where we have included the case t0=T. Further, the matrix D2w(x0,t0) is negative semidefinite. Using equation (2.1) and noting that w=Vϵ and D2w=D2Vϵ, we get, at (x0,t0),


since the matrix A, with elements Aij=δij+(p-2)uiϵujϵVϵ, is positive semidefinite. To avoid the contradiction α0, w must attain its maximum on the parabolic boundary.

Hence, for any (x,t)ΩT, we have


We finish the proof by sending α0+. ∎

With no assumptions for uϵ on the parabolic boundary, we need a stronger result, taken from [3, p. 381].

Theorem 2.5.

Let uϵ be a solution of equation (1.2), with uϵ(x,0)=u0(x). Then


Note that no condition on the lateral boundary Ω×[0,T] was used. By continuity,


for xDΩ and 0<t0tT-t0. The estimate


follows. (Here one can pass to the limit as ϵ0.)

The proof of the lemma below, a simple special case of the Miranda–Talenti lemma, can be found for smooth functions in [4, p. 308]. If f is not smooth, we perform a strictly interior approximation, so that no boundary integrals appear (which is possible since ξC0).

Lemma 2.6 (Miranda–Talenti).

Let ξC0(ΩT) and fL2(0,T,W2,2(Ω)). Then


3 Regularization

The next lemma tells us that the solutions of (1.2) converge locally uniformly to the viscosity solution of (1.1).

Lemma 3.1.

Let u be a viscosity solution of equation (1.1) and let uϵ be the classical solution of the regularized equation (1.2) with boundary values

u=uϵon parΩT.

Then uϵu uniformly on compact subsets of ΩT.


By Theorem 2.5, we can use Ascoli’s theorem to extract a convergent subsequence uϵj converging locally uniformly to some continuous function, namely, uϵjv. We claim that v is a viscosity solution of equation (1.1). The lemma then follows by uniqueness.

We demonstrate that v is a viscosity subsolution. (A symmetric proof shows that v is a viscosity supersolution.) Assume that v-ϕ attains a strict local maximum at z0=(x0,t0). Since uϵv locally uniformly, there are points


such that uϵ-ϕ attains a local maximum at zϵ. If ϕ(z0)0, then ϕ(zϵ)0 for all ϵ>0 small enough, and at zϵ, we have


Letting ϵ0, we see that v satisfies Definition 2.3 when ϕ(z0)0. If ϕ(z0)=0, let


Since |ηϵ|1, there is a subsequence such that ηϵkη when k for some ηn, with |η|1. Passing to the limit ϵk0 in equation (3.1), we see that v is a viscosity subsolution. ∎

Our proof of Theorem 1.1 consists in showing that the second derivatives D2uϵ belong locally to L2 with a bound independent of ϵ. Once this is established, we see that


Hence, for any bounded subdomain DΩT,


with C independent of ϵ. By this uniform bound, there exists a subsequence such that, as j,

(|uϵj|2+ϵj2)2-p2div((|uϵj|2+ϵj2)p-22uϵj)Uweakly in L2(D).

In particular, this means that UL2(D) and for any ϕC0(D), we have


If u is the unique viscosity solution of (1.1), we invoke Lemma 3.1 and the calculations above to find, for any test function ϕC0(D),


This shows that the Sobolev derivative ut exists and, since the previous equation holds for any subdomain DΩT, we conclude that ut=ULloc2(ΩT). To complete the proof of Theorem 1.1, it remains to establish the missing local bound of D2uϵL2 uniformly in ϵ.

4 The differentiated equation

We shall derive a fundamental identity. Let


Differentiating equation (1.2) with respect to the variable xj, we obtain


Take ξC0(ΩT), with ξ0. Multiply both sides of the equation by ξ2Vϵujϵ and sum j from 1 to n. Integrate over ΩT, using integration by parts and keeping in mind that ξ is compactly supported in ΩT, to obtain


Writing out the derivatives gives the fundamental formula


In the next section we shall bound the main term I uniformly with respect to ϵ.

5 Estimate of the second derivatives

We shall provide an estimate of the main term I. First, we record the elementary inequality


One dimension. As an exercise, we show that in this case, the second derivatives are locally bounded in L2 for any 1<p<. In one dimension, equation (1.1) reads


We absorb the terms IV and V, using Young’s inequality and inequality (5.1). For any δ>0,


Applying Theorem 2.5 we see that the right-hand side is bounded by a constant independent of ϵ>0. We have


It follows that 2uϵx2L2 locally for any p(1,).

General n. We assume for the moment that 1<p<2. We rewrite the term II involving the Laplacian as


Upon this rewriting, the term IV disappears from the equation. We focus our attention on the term involving Δ(ξuϵ). By Lemma 2.6,


Differentiating, we see that


It follows that


where f(uϵ,uiϵ,D2uϵ) depends only linearly on the second derivatives uijϵ:


By Young’s inequality, we obtain


Inserting this into the main equation gives


All terms containing D2uϵ can be absorbed by the new main term I*. To this end, we use Young’s inequality with a small parameter δ>0 to balance the terms.2 For term V, we have


Similarly, for term VII,


Using similar inequalities for the term involving f(uϵ,uϵ,D2uϵ) and choosing the parameters small enough in Young’s inequality, we find,


where C is independent of ϵ but depends on ξC2, provided that 1-54(2-p)>0, i.e., p>65. This is now a decisive restriction. Invoking Lemma 3.1 and estimate (2.2), we deduce that the majorant in (5.2) is independent of ϵ.

A symmetric proof when p>2 shows that equation (5.2) holds when p<145.

6 The case 1<p<2

In this section, we give a proof of Theorem 1.2. To this end, let ξC0(ΩT), with 0ξ1. We claim that


where the supremum norm of Vϵ=|uϵ|2+ϵ2 is taken locally, over the support of ξ. Here, uϵ is the solution of the regularized equation (1.2). This is enough to complete the proof of Theorem 1.2, in virtue of Theorem 2.5.

Multiplying the regularized equation (1.2) by (|uϵ|2+ϵ2)p-22ξ2utϵ yields


The integral of the divergence term vanishes by Gauss’s theorem and, upon integration, we have


The first integral on the right-hand side can be absorbed by the left-hand side by choosing σ=12 in


and integrating.

For the last term, the decisive observation is that


We use this in the last integral on the right-hand side to obtain


To sum up, we have now the final estimate


So far, our calculations are valid in the full range 1<p<. For 1<p<2, we have


where the supremum norm is taken over the support of ξ. Hence, equation (6.1) holds for 1<p<2 and the proof of Theorem 1.2 is complete.


We thank Amal Attouchi for valuable help with a proof.


  • [1]

    A. Attouchi, M. Parviainen and E. Ruosteenoja, C1,α regularity for the normalized p-Poisson problem, J. Math. Pures Appl. (9) 108 (2017), no. 4, 553–591.  Web of ScienceGoogle Scholar

  • [2]

    A. Banerjee and N. Garofalo, Gradient bounds and monotonicity of the energy for some nonlinear singular diffusion equations, Indiana Univ. Math. J. 62 (2013), no. 2, 699–736.  Web of ScienceCrossrefGoogle Scholar

  • [3]

    K. Does, An evolution equation involving the normalized p-Laplacian, Commun. Pure Appl. Anal. 10 (2011), no. 1, 361–396.  Web of ScienceGoogle Scholar

  • [4]

    L. Evans, Partial Differential Equations, Grad. Stud. Math. 19, American Mathematical Society, Provindence, 1998.  Google Scholar

  • [5]

    L. Evans and J. Spruck, Motion of level sets by mean curvature, J. Differential Geom. 33 (1991), no. 3, 635–681.  CrossrefGoogle Scholar

  • [6]

    , T. Jin and L. Silvestre, Hölder gradient estimates for parabolic homogeneous p-Laplacian equations, J. Math. Pures Appl. (9) 108 (2017), no. 1, 63–87.  CrossrefGoogle Scholar

  • [7]

    J. Manfredi, M. Parviainen and J. Rossi, An asymptotic mean value characterization for a class of nonlinear parabolic equations related to tug-of-war games, SIAM J. Math. Anal. 42 (2010), no. 5, 2058–2081.  CrossrefWeb of ScienceGoogle Scholar


  • 1

    Sum over repeated indices. 

  • 2

    The parameter δ is to be made so small that terms like δ0TΩξ2|D2uϵ|2𝑑x𝑑t can be moved over to the left-hand side. 

About the article

Received: 2018-04-17

Accepted: 2018-05-31

Published Online: 2018-07-12

Published in Print: 2019-03-01

Funding Source: Norges Forskningsråd

Award identifier / Grant number: 250070

Supported by the Norwegian Research Council (grant 250070).

Citation Information: Advances in Nonlinear Analysis, Volume 9, Issue 1, Pages 7–15, ISSN (Online) 2191-950X, ISSN (Print) 2191-9496, DOI: https://doi.org/10.1515/anona-2018-0091.

Export Citation

© 2020 Walter de Gruyter GmbH, Berlin/Boston. This work is licensed under the Creative Commons Attribution 4.0 Public License. BY 4.0

Comments (0)

Please log in or register to comment.
Log in