Abstract
We study a Hessian-dependent functional driven by a fully nonlinear operator. The associated Euler-Lagrange equation is a fully nonlinear mean-field game with free boundaries. Our findings include the existence of solutions to the mean-field game, together with Hölder continuity of the value function and improved integrability of the density. In addition, we prove the reduced free boundary is a set of finite perimeter. To conclude our analysis, we prove a \(\Gamma \)-convergence result for the functional.
Similar content being viewed by others
Avoid common mistakes on your manuscript.
1 Introduction
We examine Hessian-dependent functionals of the form
where \(F:S(d)\rightarrow \mathbb {R}\) is a uniformly elliptic operator, \(\Lambda >0\) is a fixed constant, \(p>d/2\), and \(S(d)\sim \mathbb {R}^\frac{d(d+1)}{2}\) stands for the space of symmetric matrices of order d. Our results include the existence of minimizers for (1), amounting to the existence of solutions to a fully nonlinear mean-field game with free boundaries. We prove Hölder-continuity of minimizers and improved integrability of the density. We also prove the free boundary has finite perimeter. Finally, we establish a result on the \(\Gamma \)-convergence of \(\mathcal {F}_{\Lambda ,p}\) and examine its consequences.
The functional in (1) is inspired by the usual one-phase Bernoulli problem, driven by the Dirichlet energy. To a limited extent, we understand \(\mathcal {F}_{\Lambda ,p}\) as a Hessian-dependent counterpart of that problem. See [2]; see also [11].
The analysis of (1) relates closely with the system
where \(F_{i,j}(M)\) denotes the derivative of F with respect to the entry \(m_{i,j}\) of M. Here, the unknown is a pair (u, m) solving the problem in a sense we make precise further.
The system in (2) amounts to the Euler-Lagrange equation associated with (1). We notice that (2) satisfies an adjoint structure. Its double-divergence equation is the formal adjoint, in the \(L^2\)-sense, of the linearized fully nonlinear problem. Due to such a distinctive pattern, we refer to (2) as a fully nonlinear mean-field game with free boundary. Interpreting the first equation in (2) as a Hamilton-Jacobi equation, it becomes natural to ask about the underlying stochastic optimal control problem. We do not examine this matter in the present paper.
Fully nonlinear elliptic operators and equations in the double-divergence form have been studied by many authors. An attempt to put together a comprehensive list of references on those topics is unrealistic. For that reason, we mention solely the monographs [8, 12] and the references therein.
Our analysis sits at the intersection of Hessian-dependent functionals, free boundary problems, and mean-field games systems. Hence we proceed with some context on those classes of problems. Hessian-dependent functionals play an essencial role in various contexts. From a purely mathematical viewpoint, they are useful to produce examples of conformally invariant energies. In dimension \(d=4\), this is the case of
whose Euler-Lagrange equation is the biharmonic operator; see [14, 15].
When it comes to applications, we mention the realm of mechanics of solids. In particular, the analysis of energy-driven pattern formation and nonlinear elasticity. For example, Hessian-dependent models play a role in studying the occurrence of wrinkles in a twisted ribbon [24]. The energy modeling the system depends on the thickness of the ribbon, denoted with h, and two symmetric tensors M and B. It has the form
Although M depends on its arguments only through lower-order terms, the tensor B depends on \(\Vert D^2u\Vert \). Another instance where Hessian-dependent functionals appear is the analysis of blister patterns in thin films on compliant substrates [6]. Here the phenomena are modeled in terms of lower-order quantities, a small Hessian-dependent perturbation, and a parameter \(h>0\). An important question concerning this class of problems is the limiting behavior \(h\rightarrow 0\); in fact, one expects that the lower and upper bounds of the functional scale similarly. We refer the reader to [18, 19, 29]. In this context, (1) amounts to an energy penalized by the measure of the positive phase. See also [4, 5, 23].
As noticed before, one can state the Euler-Lagrange equation associated with (1) in terms of the fully nonlinear mean-field game system with free boundaries (2). Mean-field games comprise a set of methods and techniques to model strategic interactions involving many players [25,26,27]; see also [28]. At the intersection of partial differential equations (PDE), stochastic analysis and numerical methods, this class of problems has attracted the attention of several authors, who have developed the theory in various directions.
Under assumptions on the stochastic dynamics governing the problem (e.g., independence of the Brownian motions among the population of players), it is possible to write a mean-field game in terms of a coupling. It comprises a Hamilton-Jacobi equation, accounting for the value of the game, and a Fokker-Planck equation describing the evolution of the population. For further references on the topics, we mention the monographs [1, 7, 13, 22]. Fully nonlinear mean-field games are the subject of [3, 16].
The interesting aspect in (2) concerns the appearance of a free boundary. At least heuristically, the game is played only in the regions where the value function is strictly positive. Combined with the free boundary condition, (2) models a game in which players optimize in the region where the value function is positive and might face extinction according to a flux condition endogenously determined.
Our first contribution is to prove the existence of solutions for the mean-field game system in (2). To that end, we impose natural assumptions on the data of the problem. Namely, we require the operator F driving the problem to be uniformly elliptic (Assumption A1) and to satisfy a norm-like growth condition (Assumption A3). We also require the operator F to be convex (Assumption (A2) and the boundary data g to be a function in \(W^{2,p}(B_1)\) (Assumption A4). We report our findings in the following theorem.
Theorem 1
(Existence and regularity of solutions) Suppose Assumptions A1, A2, A3, and A4, to be detailed further, are in force. In addition, suppose \(p<d<2p\). Then there exists a solution \((u,m)\in W^{2,p}(B_1)\times L^1(B_1)\) to (2). Also, for every \(\alpha \in (0,p/d)\), we have \(u\in C^{\alpha }_{\textrm{loc}}(B_1)\) and there exists \(C_\alpha >0\) such that
The constant \(C_\alpha >0\) depends on the exponent \(\alpha \).
If, in addition, F is strictly convex and \(p>2\), we can prove that m is not only integrable but is indeed an \(L^\frac{p}{p-1}\)-function, with estimates; c.f. [21]. To establish Theorem 1, we start with the direct method in the calculus of variations and the existence of minimizers for (1). Then we turn our attention to (2). First, we resort to the theory of weak solutions available for equations in the double-divergence form. Finally, elements in the \(L^p\)-viscosity theory lead to the existence of solutions to the system. To complete the proof, we resort to a delicate application of Sobolev inequalities.
Once we have established the existence of solutions for (2) and produced a regularity result, we examine the free boundary. Regularity results for the solutions build upon ingredients of geometric measure theory to ensure the reduced free boundary is a set of finite perimeter. We summarize our findings in this direction in the following result.
Theorem 2
(Free boundary condition and finite perimeter) Let \(u\in W^{2,p}_{\textrm{loc}}(B_1)\cap W^{1,p}_g(B_1)\) be a local minimizer for (1), for \(p>d/2\). Suppose Assumptions A1, A2, A3 and A4, to be detailed further, are in force. Then the reduced free boundary, denoted with \(\partial ^*\{u>0\}\), is a set of finite perimeter.
The remainder of this paper is organized as follows. Section 2.1 details our main assumptions, whereas Sect. 2.2 gathers preliminary material and results. Section 3 presents the proof of Theorem 1. In Sect. 4, we examine the free boundary and put forward the proof of Theorem 2. A final section closes the paper with a \(\Gamma \)-convergence result and some consequences.
2 Preliminaries
This section presents the main assumptions under which we work and collects some preliminary notions and results.
2.1 Main Assumptions
We proceed with a condition on the uniform ellipticity of the operator F.
A 1
(Uniform ellipticity) We suppose the operator \(F: S(d) \rightarrow \) \(\mathbb {R}\) is \(\lambda \)-elliptic for some \(\lambda \ge 1\). That is,
for every \(M, N \in S(d)\), with \(N \ge 0 \). We also suppose \(F(0)=0\). Finally, we require \(F_{ij}(M)=F_{ji}(M)\), for every \(i,j=1,\ldots ,d\), where \(F_{ij}(M)\) stands for the derivative of F with respect to the entry (i, j) of M.
Remark 1
Note that A1 implies a coercivity condition on F over non-negative matrices. By taking \(M \equiv 0\), the inequalities in (3) yield
for every \(N \ge 0 \).
Next, we impose a convexity condition on the operator F.
A 2
(Convexity of the operator F) We suppose the operator \(F=F(M)\) to be convex with respect to M.
Part of our arguments requires F to satisfy a coercivity condition in the entire S(d). To that end, we strength A1 as follows.
A 3
(Growth condition) We suppose there exists \(\lambda \ge 1\) such that the operator F satisfies
for every \(M \in S(d)\).
The typical example of an operator \(F=F(M)\), satisfying the former assumptions, depends on M through its norm. For instance, let \(A\in \mathbb {R}\) be a constant and consider
For a more general operator, including explicit dependence on the space-variable \(x\in B_1\), we consider \(A\in W^{2,p}_{\textrm{loc}}(B_1)\) and define \(F=F(x,M)\) as
For further examples related to the previous one see [6, 23].
Notice that A3 implies \(F\ge 0\); this fact is important when studying the weak lower semicontinuity of our functional. We conclude this section with an assumption on the boundary data.
A 4
(Boundary data) We suppose the function \(g\in W^{2,p}(B_1)\) is non-negative and non-trivial.
2.2 Preliminary Notions and Results
For the sake of completeness, we recall definitions and former results we use throughout the manuscript. We continue with a definition
Definition 1
(Affine Sobolev spaces) Let \(p>d/2\) and \(g\in W^{2,p}_{\textrm{loc}}(B_1)\). We say that
if \(u\in W^{2,p}_{\textrm{loc}}(B_1)\) and \(u-g\in W^{1,p}_0(B_1)\).
From a PDE perspective, having \(u\in W^{2,p}_{\textrm{loc}}(B_1)\cap W^{1,p}_g(B_1)\) is tantamount to prescribe \(u=g\) on \(\partial B_1\) in the Sobolev sense. This interpretation will be helpful when relating (1) and (2).
As usual in the literature on mean-field games [25,26,27], a solution to (2) relies on two distinct definitions – namely, the notions of viscosity and weak (distributional) solutions. We proceed by recalling the definition of \(L^p\)-viscosity solution of a fully nonlinear elliptic equation; see [10, Definition 2.1].
Definition 2
(\(L^p\)-viscosity solutions) Let \(F:S(d)\rightarrow \mathbb {R}\) be a fully nonlinear operator satisfying A1 and \(f\in L^p_{\textrm{loc}}(B_1)\), for \(p>d/2\). A function \(u \in C(B_1)\) is an \(L^{p}\)-viscosity sub-solution (resp. super-solution) of
if, for all \(\varphi \in W_{\textrm{loc}}^{2, p}(B_1)\), whenever \(\varepsilon >0, U \subset B_1\) is open, and
then \(u-\varphi \) cannot have a local maximum (resp. minimum) in U. Moreover, if u is both an \(L^p\)-viscosity sub-solution and an \(L^p\)-viscosity super-solution, u is said to be an \(L^p\)-viscosity solution.
The definition of \(L^p\)-viscosity solution is necessary since \(L^p\)-functions might not be defined at the points where the usual conditions must be tested. For a comprehensive account of this notion, we refer the reader to [10]. We continue with the definition of weak solutions for double-divergence equations.
Definition 3
(Weak solution) Let \(A\in L^\infty (B_1,S(d))\) and denote \(A(x)=:[a_{i,j}(x)]_{i,j=1}^d\). Suppose
We say \(m\in L^1(B_1)\) is a weak solution to
if, for every \(\phi \in C^\infty _c(B_1)\) we have
A solution to the mean-field game in (2) combines Definitions 2 and 3.
Definition 4
(Solution for the MFG system) The pair (u, m) is a weak solution to (2) if the following hold:
-
1.
We have \(u\in C(B_1)\cap W^{1,p}_g\) and \(m\in L^1(B_1)\), with \(m\ge 0\);
-
2.
The function u is an \(L^p\)-viscosity solution to
$$\begin{aligned} F(D^2u)=m^\frac{1}{p-1}\hspace{.2in}\text{ in }\hspace{.2in}B_1\cap \{u>0\}; \end{aligned}$$ -
3.
The function m is a weak solution to
$$\begin{aligned} \left( F_{ij}(D^2u)m\right) _{x_ix_j}=0\hspace{.2in}\text{ in }\hspace{.2in}B_1\cap \{u>0\}. \end{aligned}$$
Next, we recall the Poincaré’s inequality for functions lacking compact support. In particular, we are interested in \(u\in W^{1,p}_g(B_1)\).
Lemma 1
(Poincaré’s inequality) Let \(u \in W_{g}^{1, p}\left( B_{1}\right) \) and \(C_{p}>0\) be the Poincaré’s constant associated with \(L^{p}\left( B_{1}\right) \) and the dimension d. Then for every \(C<C_{p}\), there exists \(C_{1}\left( C, C_{p}\right) >0\) and \(C_{2} \ge 0\) such that
For the detailed proof of this fact, we refer the reader to [20, Lemma 2.7, p. 22]. It follows from \(u-g\in W^{1,p}_0(B_1)\) and the usual Poincaré’s inequality.
In Sect. 3, we deal with the existence of minimizers for \(\mathcal {F}_{\Lambda ,p}\) in \(W^{2,p}_\textrm{loc}(B_1)\cap W^{1,p}_g(B_1)\). Our reasoning uses the weak lower-semicontinuity of the functional
this is the content of the following lemma.
Lemma 2
Let \(p>d/2\) and suppose A2, A3 and A4 hold true. Let \(\left( u_{n}\right) _{n \in \mathbb {N}} \subset W^{2,p}_{\textrm{loc}}(B_1)\cap W^{1,p}_g(B_1)\) be such that
Then,
For the proof of Lemma 2, we refer to [3, Proposition 3]. In what follows, we detail the proof of Theorem 1.
3 Existence of Solutions
In this section, we present the proof of Theorem 1; we start by establishing the existence of minimizers for (1).
Proposition 1
(Existence of minimizers) Suppose Assumptions A2, A3, and A4 are in force and fix \( p>d/2\), arbitrary. Then there exists \(u^*\in W^{2,p}_\textrm{loc}(B_1)\cap W^{1,p}_g(B_1)\) such that
for all \(u\in W^{2,p}_{\textrm{loc}}(B_1)\cap W^{1,p}_g(B_1)\).
Proof
Under Assumptions A2 and A3, the existence of minimizers follows from the direct method in the calculus of variations. We split the argument into three steps.
Step 1 - We first examine
In view of the Remark 1, \(\gamma \ge 0\). Furthermore, since \(g\in W^{2,p}(B_1)\),
Hence, \(0\le \gamma \le C(g,\Lambda )<\infty \). Let \(\left( u_{n}\right) _{n \in \mathbb {N}} \subset W^{2,p}_\textrm{loc}(B_1)\cap W^{1,p}_g(B_1)\) be a minimizing sequence; there exists \(N \in \mathbb {N}\) such that
for every \(n\ge N\). Therefore, for all \(n\ge N\),
In the next step the upper bound for \(D^2u_n\) builds upon properties of the functional.
Step 2 - As a consequence of the former inequality, we infer that \(\left( D^{2} u_{n}\right) _{n \in \mathbb {N}}\) is uniformly bounded in \(L^{p}\left( B_{1}\right) \). Since \(p>d/2\), the embedding \(W^{2,p}(B_1)\hookrightarrow W^{1,p}(B_1)\) is compact. Furthermore, we conclude that \(\left( u_{n}\right) _{n \in \mathbb {N}}\) is uniformly bounded in \(W^{2,p}_{\textrm{loc}}(B_1)\cap W^{1,p}_g(B_1)\); it follows from Lemma 1 combined with general facts [17]. Hence, there exists \(u_{\infty } \in W^{2,p}_{\textrm{loc}}(B_1)\cap W^{1,p}_g(B_1)\) such that
and
The result follows at once if we ensure that
and
hold. Notice that Lemma 2 combines the convergence mode in (4) to yield (6). In the sequel, we establish (7).
Step 3 - Because of the strong convergence (5), there exists a subsequence, also denoted with \((u_n)_{n\in \mathbb {N}}\), and a negligible subset \(\mathcal {N}\subset B_1\), such that \(u_n(x)\rightarrow u_\infty (x)\) for every x in \(B_1\setminus \mathcal {N}\). As a consequence, if \(u_\infty (x)>0\), there exists \(N\in \mathbb {N}\) such that \(u_n(x)>0\) for every \(n\ge N\). If \(u_\infty (x)=0\), then \(\chi _{\{u_\infty >0\}}(x)=0\). Therefore,
for almost every \(x\in B_1\setminus \mathcal {N}\). Hence,
which completes the proof. \(\square \)
Remark 2
We notice the minimizing sequence \((u_n)_{n\in \mathbb {N}}\) is uniformly bounded in \(W^{2,p}(B_1)\). As a consequence, it is also uniformly bounded in some Hölder space. Therefore, we could have used uniform convergence in (4).
We close this section with the proof of Theorem 1.
Proof of Theorem 1
We split the proof into four steps.
Step 1 - Let \(u^*\in W^{2,p}_{\textrm{loc}}(B_1)\cap W^{1,p}_g(B_1)\) be the minimizer for (1) whose existence follows from Proposition 1. There exists \(\mathcal {N}\subset B_1\) such that \(D^2u^*(x)\) is well-defined for every \(x\in B_1\setminus \mathcal {N}\), with \(|\mathcal {N}|=0\). This fact, combined with A3, implies that \(F(D^2u^*(x))\ge 0\) for almost every \(x\in B_1\). Therefore, \(u^*\) satisfies \(F(D^2u^*)\ge 0\) in the \(L^p\)-viscosity sense; see [10, Lemma 2.6].
Step 2 - By considering a variation of \(u^*\) compactly supported in \(B_1\cap \{u>0\}\), we obtain
for every \(\varphi \in C^\infty _c(B_1\cap \{u>0\})\). Set \(F(D^2u^*)=:m^\frac{1}{p-1}\); we infer that m(x) is well-defined and satisfies \(m(x)\ge 0 \) for almost every \(x\in B_1\). In addition,
that is, \(m\in L^1(B_1)\). Finally, we notice the integral in (9) is well-defined and leads to
for every \(\varphi \in C^\infty _c(B_1\cap \{u>0\})\).
Step 3 - It remains to check that \(u^*\) is an \(L^p\)-viscosity solution to the first equation in (2). The definition of m implies that \(u^*\) satisfies
for almost every \(x\in B_1\cap \{u^*>0\}\). As before, an application of [10, Lemma 2.6] ends the proof.
Step 4 - We prove that \(Du^*\in L^r(B_1)\) for every \(d<r<dp/(d-p)\). We start by recalling the Gagliardo-Nirenberg inequality for bounded domains. Being \(u^*\in W^{2,p}_\textrm{loc}(B_1)\cap W^{1,p}_{g}(B_1)\) a minimizer for (1), there exists \(C_1,C_2>0\) such that
provided
for some \(1/2<\alpha <1\) and \(q_2>0\). We notice the \(L^p\)-norm of \(D^2g\) appears in (11) because
because of Assumption A3 and the fact that g is a competitor for \(u^*\).
Given \(d\ge 2\), \(p>d/2\), and \(1<r<\infty \). it is always possible to find \(\alpha \in (1/2,1)\) and \(q_1>1\) such that (11) is satisfied. Because \(F(D^2u^*)\ge 0\), we know that for every \(q>0\) there exists \(C>0\) such that
see [12, Theorem 4.8, item (2)]. Hence, (10) becomes
and a straightforward application of Morrey’s Theorem completes the proof. \(\square \)
Remark 3
(Improved integrability for m)Let (u, m) be a weak solution to the fully nonlinear MFG (2). In case F is strictly convex and \(p>2\), we claim that \(m\in L^\frac{p}{p-1}(B_1)\). In fact, m is defined almost everywhere in \(B_1\) as \(m=F(D^2u)^{p-1}\). Under the strict convexity of F and \(p>2\), solutions to the Euler-Lagrange equation are minimizers for the functional (1). Hence, A3 transmits the integrability of \(D^2u\in L^p(B_1)\) to m, and the claim follows. Compare with [21]; see also [8]. Re-writing the exponent above as \(1+1/(p-1)\) we quantify the improved integrability of m in face of the \(L^1\)-regime.
Remark 4
(Improved regularity for the value function)The value function is \(\alpha \)-Hölder-continuous, for every \(\alpha \in (0,1)\). Hence, the regularity established in the former argument amounts to an improvement of the usual Krylov-Safonov regularity theory implied by uniform ellipticity.
4 Information on the Free Boundary
In the sequel, we examine local properties of the free boundary \(\partial \{u>0\}\) and present the proof of Theorem 2. The following corollary connects the regularity of minimizers with information on the free boundary. We refer to it when proving the first part of Theorem 2.
Corollary 1
Let \(x_0\in B_1\) and \(0<r<\textrm{dist}(x_0,\partial B_1)\). Suppose that \(u\in W^{2,p}_{\textrm{loc}}(B_r(x_0))\) is non-negative and satisfies the following minimality condition: Given \(p>d/2\),
for every \(v\in W^{2,p}_{\textrm{loc}}(B_r(x_0))\) such that
Assume also that A1–A4 holds true. There exists \(\varepsilon _0>0\) such that, for every \(0<\varepsilon \le \varepsilon _0\) one finds a universal constant \(C>0\) for which
Proof
We split the argument into four steps and begin by proving that, for given \(0<\varepsilon \le \varepsilon _0\) (fixed and to be chosen later), one gets
for some universal constant \(C>0\)
Step 1 - We begin by fixing a function \(\psi \in \mathcal {C}^{\infty }(\mathbb {R}^d)\) such that
and
for some universal constant \(C_1>0\). For a fixed \(0<\varepsilon _0\), consider the functions
and
which, by the minimality condition (12), give
Step 2 - Now we calculate \(F(D^2\tilde{u}_{\varepsilon })^p\) in \(B_r(x_0)\); to do so, first notice that \(\tilde{u}_{\varepsilon _0}\) can be writen as
Therefore,
where \(M:=Du^TD\psi +DuD\psi ^T\). Combining the former equality with Assumption A3 one gets
In the next step we detail an involved chain of inequalities used in the argument.
Step 3 - We combine (16) and (15), set
and resort to Assumption A3 to compute
Therefore
where
and
The argument in the proof of Theorem 1 ensures that
Therefore, there exists a universal constant \(C_2>0\) such that
Recall that \(M=Du^TD\psi +DuD\psi ^T\). Because of the bounds imposed on \(\psi \) and the estimates available for Du, we conclude
Hence, by requiring \(\varepsilon _0<1\), the Hölder inequality and the Theorem 1 yield
where \(C_3\) and \(C_4\) are positive, universal constants. Hence, there exists a universal constant \(C_5>0\) such that
Thus one finds \(C_6>0\) a universal constant for which
By combining (17), (18) and (19) we have that
Finally, set \(\varepsilon _0=O(\Vert Du\Vert _{L^p})\). Theorem 1 ensures that for every \(0<\varepsilon <\varepsilon _0\) one obtains
where the constant \(C>0\) is now universal.
Step 4 - Now, Assumption A3 yields
To estimate \(\Vert Du\Vert _{L^p(\{0<u\le \varepsilon \}\cap B_{r/2}(x_0))}\) we recall the Galiardo-Nirenberg inequality for bounded domains. If \(u\in W^{2,p}_{\textrm{loc}}(B_1)\), there exists universal constants \(C_4,C_5>0\) such that
provided
for some \(1/2<\alpha <1\) and \(q_2>0\), where \(\Gamma _\varepsilon :=\{0<u\le \varepsilon \}\cap B_{r/2}(x_0))\). Given \(p>d/2\) and \(d\ge 2\) it is always possible to find \(\alpha \in (1/2,1)\) and \(q_1>p\) satisfying (21), which implies that there exists \(C_6:=C_6(\alpha ,p,d,\Vert u\Vert _{L^p(\Gamma _\varepsilon )})>0\) such that
Also,
By combining the former inequalities, we get
where \({\tilde{C}}:=\tilde{C}(n,p,\alpha ,\lambda ,C_1,\Vert u\Vert _{L^p(\Gamma _\varepsilon )})>0\) is an universal constant. A straightforward application of the area formula yields
and finishes the proof. \(\square \)
4.1 Proof of Theorem 2
In what follows, we organize the previous results and present the proof of Theorem 2. The Sobolev regularity of minimizers and its corollary leads to the finite perimeter of the reduced free boundary.
Proof of Theorem 2
Because of Corollary 1, there exists a sequence \((\delta _n)_{n\in \mathbb {N}}\subset \mathbb {R}\) of real numbers, with \(\delta _n\rightarrow 0\), satisfying
for every \(n\in \mathbb {N}\). Standard convergence results ensure that
Finally, the lower semi-continuity of the perimeter implies
and yields the conclusion. \(\square \)
5 Perturbation Analysis via \(\Gamma \)-Convergence
This section specializes the operator F to be the norm and considers small values of the parameter \(\Lambda \) in (1). We regard the functional
as a free boundary perturbation of
Denote with \(u_\Lambda \) a minimizer for (22) and with \(u_0\) the minimizer for (23). We are interested in the behavior of \((u_\Lambda )_{\Lambda >0}\), as \(\Lambda \rightarrow 0\). In particular, we search for the topologies where the convergence \(u_\Lambda \rightarrow u_0\) is available. Our starting point is a \(\Gamma \)-convergence result. Namely, we first prove that \(\mathcal {G}_{\Lambda ,p}\xrightarrow {\Gamma }\mathcal {G}_{0,p}\) as \(\Lambda \rightarrow 0\).
Although interesting on its own merits, the \(\Gamma \)-convergence problem is motivated by its potential consequences on the regularity theory of minimizers to (22). Indeed, we use properties of \(\Gamma \)-convergence to prove an approximation result. It states that minimizers are close, in a suitable topology, to a minimizer of the \(\Gamma \)-limit (see Proposition 2). This type of approximation result is central to perturbative methods in regularity theory; see, for instance, [9, 12]. We believe the \(\Gamma \)-convergence analysis can be used as an ingredient in the study of improved regularity for Hessian-dependent functionals through approximation methods. We proceed with some auxiliary lemmas.
Lemma 3
(Equicoerciveness) Let \(p>1\) be fixed and \((\Lambda _n)_{n\in \mathbb {N}}\) be a sequence such that \(\Lambda _n\rightarrow 0\), as \(n\rightarrow \infty \). Define the functional \(\mathcal {G}_{n,p}:L^p(B_1)\rightarrow \mathbb {R}\) as
if \(v\in W^{2,p}(B_1)\), and \(\mathcal {G}_{n,p}[v]:=+\infty \) in case \(v\in L^p(B_1)\setminus W^{2,p}(B_1)\). Let \((u_m)_{m\in \mathbb {N}}\subset L^p(B_1)\) be such that
for every \(m\in \mathbb {N}\) and some \(C>0\). Then \(\Vert u_m\Vert _{W^{2,p}(B_1)}\le C\), uniformly in \(m\in \mathbb {N}\), for some \(C>0\).
Proof
It follows from (24) that
By Lemma 1 and standard inequalities available for Sobolev spaces [17], there exists \(C>0\) such that
uniformly in \(m\in \mathbb {N}\). \(\square \)
Before continuing, we introduce the functional \(\mathcal {G}_{0,p}:L^p(B_1)\rightarrow \mathbb {R}\), given by
if \(v\in W^{2,p}(B_1)\), and \(\mathcal {G}_{0,p}[v]:=+\infty \) if \(v\in L^p(B_1){\setminus } W^{2,p}(B_1)\). The next lemma relates \(\mathcal {G}_{n,p}\) and \(\mathcal {G}_{0,p}\).
Lemma 4
Let \(p>1\) be fixed and \((\Lambda _n)_{n\in \mathbb {N}}\) be a sequence of real numbers so that \(\Lambda _n\rightarrow 0\), as \(n\rightarrow \infty \). For each \(u\in L^p(B_1)\) there exists a sequence \((u_n)_{n\in \mathbb {N}}\in L^p(B_1)\) converging strongly to u in \(L^p(B_1)\), such that
Proof
Let \(u\in L^p(B_1)\) be given and \(u_n:=u\), for every \(n\in \mathbb {N}\). If \(u\in L^p(B_1)\setminus W^{2,p}(B_1)\), we get
and (25) is immediately satisfied. Conversely, suppose \(u\in W^{2,p}(B_1)\). In that case, we have
\(\square \)
Lemma 5
Let \(p>1\) be fixed and \((\Lambda _n)_{n\in \mathbb {N}}\) be a sequence of real numbers so that \(\Lambda _n\rightarrow 0\), as \(n\rightarrow \infty \). Given \((u_n)_{n\in \mathbb {N}}\subset L^p(B_1)\) and \(u\in L^p(B_1)\), with \(u_n\rightarrow u\) strongly in \(L^p(B_1)\), we have
Proof
To deduce (26) from the strong convergence, suppose first \((u_n)_{n\in \mathbb {N}}\subset L^p(B_1){\setminus } W^{2,p}(B_1)\). Then
and (26) follows. Otherwise, suppose \((u_n)_{n\in \mathbb {N}}\subset W^{2,p}(B_1)\).
Through a subsequence, if necessary, we can suppose the \(\liminf \) in (26) is in fact a limit. If such a limit is not finite, then (26) trivially holds. Suppose otherwise; if this limit is finite, there exists \(C>0\) such that
for every \(n\in \mathbb {N}\), large enough (and therefore for every \(n\in \mathbb {N}\)). As a consequence, \(\Vert D^2u_n\Vert _{L^p(B_1)}\) is uniformly bounded; evoking once again standard inequalities for Sobolev functions, one infers the existence of a constant \(C>0\) such that
The weakly lower semi-continuity of the \(L^p\)-norm yields
and completes the proof. \(\square \)
By combining Lemmas 3, 4, and 5, we derive the following theorem.
Theorem 3
(Gamma Convergence) Let \(p>d/2\) be fixed and \((\Lambda _n)_{n\in \mathbb {N}}\) be a sequence of real numbers so that \(\Lambda _n\rightarrow 0\), as \(n\rightarrow \infty \). Then \(\mathcal {G}_{n,p}\xrightarrow {\Gamma }\mathcal {G}_{0,p}\).
In the sequel, we explore a consequence of the \(\Gamma \)-convergence result. It consists of an approximation result by \(C^{1,\alpha }\)-regular functions.
5.1 Regular Approximations
We have proved that minimizers for (1) are Hölder-continuous. However, the use of \(\Gamma \)-convergence allows us to arbitrarily approximate minimizers by \(C^{1,\alpha }\)-regular functions. This is the content of the following proposition
Proposition 2
(\(C^{1,\alpha }\)-approximation) Let \(p>d/2\) be fixed. Given \(\delta >0\), there exists \(\varepsilon >0\) such that, if \(\Lambda <\varepsilon \) and \(u\in W^{2,p}(B_1)\cap W^{1,p}_g(B_1)\) be a minimizer for (1), one can find \(h\in C^{1,\alpha }_{\textrm{loc}}(B_1)\) satisfying
Proof
We use a contradiction argument. Suppose the statement of the proposition is false. In this case, there exist a real number \(\delta _0>0\) and sequences \((u_n)_{n\in \mathbb {N}}\) and \((\Lambda _n)_{n\in \mathbb {N}}\) such that
as \(n\rightarrow \infty \),
for every \(v\in W^{2,p}_{\textrm{loc}}(B_1)\cap W^{1,p}_g(B_1)\) and every \(n\in \mathbb {N}\), but
for every \(h\in C^{1,\alpha }_{\textrm{loc}}(B_1)\), and every \(n\in \mathbb {N}\).
However,
for some \(C>0\). Hence, there exists \(u_\infty \in W^{2,p}(B_1)\cap W^{1,p}_g(B_1)\) such that \(u_n\) converges \(u_\infty \), weakly in \(W^{2,p}(B_1)\) and strongly in \(W^{1,p}_g(B_1)\). That is tantamount to say that \(u_\infty \) is an accumulation point for the sequence \((u_n)_{n\in \mathbb {N}}\).
Because of Theorem 3, we conclude that \(u_\infty \) is a minimizer for \(\mathcal {G}_{0,p}\). Previous results in the literature ensure that \(u_\infty \in C^{1,\alpha }_\textrm{loc}(B_1)\) [3]. By taking \(h:=u_\infty \) in (27), we get a contradiction and complete the proof.\(\square \)
Data Availability
Not applicable.
References
Achdou, Y., Cardaliaguet, P., Delarue, F., Porretta, A., Santambrogio, F.: Mean field games, volume 2281 of Lecture Notes in Mathematics. In: P. Cardaliaguet, A. Porretta (eds). Centro Internazionale Matematico Estivo (C.I.M.E.), Florence [2020] (2020). Fondazione CIME/CIME Foundation Subseries. Springer, Cham
Alt, H.W., Caffarelli, L.A.: Existence and regularity for a minimum problem with free boundary. J. Reine Angew. Math. 325, 105–144 (1981)
Andrade, P., Pimentel, E.A.: Stationary fully nonlinear mean-field games. J. Anal. Math. 145(1), 335–356 (2021)
Aviles, P., Giga, Y.: A mathematical problem related to the physical theory of liquid crystal configurations. In: Miniconference on geometry and partial differential equations, 2 (Canberra, 1986), volume 12 of Proc. Centre Math. Anal. Austral. Nat. Univ., pages 1–16. Austral. Nat. Univ., Canberra (1987)
Aviles, P., Giga, Y.: The distance function and defect energy. Proc. Roy. Soc. Edinburgh Sect. A 126(5), 923–938 (1996)
Bedrossian, J., Kohn, R.V.: Blister patterns and energy minimization in compressed thin films on compliant substrates. Comm. Pure Appl. Math. 68(3), 472–510 (2015)
Bensoussan, A., Frehse, J., Yam, P.: Mean field games and mean field type control theory, Springer Briefs in Mathematics. Springer, New York (2013)
Bogachev, V.I., Krylov, N.V., Röckner, M., Shaposhnikov, S.V.: Fokker-Planck-Kolmogorov equations. Mathematical Surveys and Monographs, vol. 207. American Mathematical Society, Providence, RI (2015)
Caffarelli, L.: Interior a priori estimates for solutions of fully nonlinear equations. Ann. Math. (2) 130(1), 189–213 (1989)
Caffarelli, L., Crandall, M.G., Kocan, M., Święch, A.: On viscosity solutions of fully nonlinear equations with measurable ingredients. Comm. Pure Appl. Math. 49(4), 365–397 (1996)
Caffarelli, L., Salsa, S.: A geometric approach to free boundary problems. Graduate Studies in Mathematics, vol. 68. American Mathematical Society, Providence, RI (2005)
Caffarelli, L.A., Cabré, X.: Fully nonlinear elliptic equations American Mathematical Society Colloquium Publications, vol. 43. American Mathematical Society, Providence, RI (1995)
Cardaliaguet, P.: Notes on mean-field games (2013)
Chang, S.-Y.A., Gursky, M.J., Yang, P.C.: Regularity of a fourth order nonlinear PDE with critical exponent. Amer. J. Math. 121(2), 215–257 (1999)
Chang, S.-Y.A., Wang, L., Yang, P.C.: A regularity theory of biharmonic maps. Comm. Pure Appl. Math. 52(9), 1113–1137 (1999)
Chowdhury, I., Jakobsen, E.R., Krupski, M.: On fully nonlinear parabolic mean field games with examples of nonlocal and local diffusions (2021)
Cianchi, A., Maz’ya, V.: Sobolev inequalities in arbitrary domains. Adv. Math. 293, 644–696 (2016)
Conti, S., Maggi, F.: Confining thin elastic sheets and folding paper. Arch. Ration. Mech. Anal. 187(1), 1–48 (2008)
Conti, S., Maggi, F., Müller, S.: Rigorous derivation of Föppl’s theory for clamped elastic membranes leads to relaxation. SIAM J. Math. Anal. 38(2), 657–680 (2006)
Dal Maso, G.: An introduction to \(\Gamma \)-convergence. Progress in Nonlinear Differential Equations and their Applications, vol. 8. Birkhäuser Boston Inc, Boston, MA (1993)
Fabes, E.B., Stroock, D.W.: The \(L^p\)-integrability of Green’s functions and fundamental solutions for elliptic and parabolic equations. Duke Math. J. 51(4), 997–1016 (1984)
Gomes, D., Pimentel, E., Voskanyan, V.: Regularity theory for mean-field game systems. Springer Briefs in Mathematics. Springer, Cham (2016)
Kohn, R.V.: Energy-driven pattern formation. In: International Congress of Mathematicians. Vol. I, pages 359–383. Eur. Math. Soc., Zürich (2007)
Kohn, R.V., O’Brien, E.: The wrinkling of a twisted ribbon. J. Nonlinear Sci. 28(4), 1221–1249 (2018)
Lasry, J.-M., Lions, P.-L.: Jeux à champ moyen. I. Le cas stationnaire. C. R. Math. Acad. Sci. Paris 343(9), 619–625 (2006)
Lasry, J.-M., Lions, P.-L.: Jeux à champ moyen. II. Horizon fini et contrôle optimal. C. R. Math. Acad. Sci. Paris 343(10), 679–684 (2006)
Lasry, J.-M., Lions, P.-L.: Mean field games. Jpn. J. Math. 2(1), 229–260 (2007)
Lions, P.-L.: Cours au collège de france. www.college-de-france.fr
Venkataramani, S.: Lower bounds for the energy in a crumpled elastic sheet-a minimal ridge. Nonlinearity 17(1), 301–312 (2004)
Acknowledgements
The authors are grateful to Giovanni Bellettini for his comments on the material in this paper. Partially supported by the Centre for Mathematics of the University of Coimbra (funded by the Portuguese Government through FCT/MCTES, https://doi.org/10.54499/UIDB/00324/2020). JC is funded by FAPERJ-Brazil (# E26/202.075/2020). EP is partly funded by FAPERJ-Brazil (Grant # E26/200.002/2018), ICTP-Trieste and Instituto Serrapilheira (Grant # 1811-25904). This study was financed in part by the Coordenação de Aperfeiçoamento de Pessoal de Nível Superior - Brazil (CAPES) - Finance Code 001.
Funding
Open access funding provided by FCT|FCCN (b-on).
Author information
Authors and Affiliations
Corresponding author
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.
About this article
Cite this article
Correa, J.C., Pimentel, E.A. A Hessian-Dependent Functional With Free Boundaries and Applications to Mean-Field Games. J Geom Anal 34, 95 (2024). https://doi.org/10.1007/s12220-023-01542-0
Received:
Accepted:
Published:
DOI: https://doi.org/10.1007/s12220-023-01542-0