Uniform estimates and Brezis-Merle type inequalities for the k k -Hessian equation
Jie Deng, Haibin Wang, Bin Zhou · 2026
Read Uniform estimates and Brezis-Merle type inequalities for the k k -Hessian equation by Jie Deng, Haibin Wang, Bin Zhou (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2603-25511.txt.
Introduction
Deng, Wang, and Zhou prove a Brezis-Merle style bound for the k-Hessian equation. The bound governs k-convex functions that vanish at the border of their domain. As a payoff, they get an Alexandrov-Bakelman-Pucci estimate for the mid-level Hessian equation. That kind of estimate caps a solution by the size of its source term. They also build a concentration-compactness rule for how solutions of the mean-field k-Hessian equation can blow up. The tools serve a hard class of fully nonlinear equations.
Catalog record
| CREATOR | Jie Deng, Haibin Wang, Bin Zhou |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 26814 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2603.25511
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Uniform estimates and Brezis-Merle type inequalities for the k k -Hessian equation Thanks: This research is partially supported by National Key R & \& D Program of China 2023YFA009900 and NSFC Grant 12271008. Jie Deng, Haibin Wang, Bin Zhou Abstract . In this paper, we prove a Brezis-Merle type inequality for k k -convex functions vanishing on the boundary. As an application, we establish an Alexandrov-Bakelman-Pucci type estimate for the intermediate Hessian equation. Furthermore, we establish a concentration-compactness principle for the blow-up behavior of solutions to the Liouville type k k -Hessian equations. 1. Introduction Let \Omega be a bounded smooth domain in {\bf\mathbb{R}}^{n} . For u\in C^{2}(\Omega) , k=1,\dots,n , the k -Hessian operator S_{k} is \displaystyle S_{k}[u]=\sum_{1\leq i_{1}<...<i_{k}\leq n}\lambda_{i_{1}}\cdots\lambda_{i_{k}}, where \lambda_{1},...,\lambda_{n} are the eigenvalues of the Hessian matrix D^{2}u . When k=1 , it is the Laplacian operator. Let \Phi^{k}(\Omega)=\{u\in C^{2}(\Omega)\,|\,S_{j}[u]\geq 0,\forall j=1,...,k\} be the sets of smooth k -convex functions. k -convex functions are admissible functions for S_{k} to be elliptic. The notion of k -convexity can be extended to general upper semi-continuous functions. An upper semi-continuous function u:\Omega\to[-\infty,+\infty) is called k -convex in \Omega if S_{k}[q]\geq 0 for all quadratic polynomials q for which the difference u-q has a finite local maximum in \Omega . In a series of papers [ 13 , 14 , 15 ] , Trudinger-Wang introduced the Hessian measures for general k -convex functions and developed the weak solution theory for the k -Hessian equation (1.1) S_{k}[u]=f. A key property shown in [ 14 ] is the weak continuity of S_{k}[u] . Namely, if there is a sequence \{u_{m}\} in \Phi^{k}(\Omega) converging locally in L^{1}_{\text{loc}}(\Omega) to u and S_{k}[u_{m}] converges weakly to the k -Hessian measure \mu_{k}[u] in the sense of measures. The uniform estimate for the k -Hessian equations was established by [ 5 ] under the assumption that f\in L^{p}(\Omega) where p>\frac{n}{2k} if k\leq\frac{n}{2} or p=1 if k>\frac{n}{2} , based on the Hessian Sobolev inequalities [ 17 , 18 ] . In a famous paper [ 4 ] , the uniform estimates and blow-up behavior for equation -\triangle u=V(x)e^{u} were investigated based on the following basic inequality: \int_{\Omega}\exp\left[\frac{(4\pi-\delta)|u(x)|}{\|\triangle u\|_{L^{1}(\Omega)}}\right]\,dx\leq\frac{4\pi^{2}}{\delta}(\mathrm{diam}(\Omega))^{2} for all \delta\in(0,4\pi) and u on a bounded domain \Omega\subset\mathbb{R}^{2} with vanishing boundary value. Subsequently, in higher dimensions, Aguilar and Peral [ 1 ] established the inequality \int_{\Omega}\exp\left[\frac{(n^{\frac{n}{n-1}}\omega_{n}^{\frac{1}{n-1}}-\delta)|u(x)|}{\|f\|^{\frac{1}{n-1}}_{L^{1}(\Omega)}}\right]\,dx\leq\frac{n^{\frac{n}{n-1}}\omega_{n}^{\frac{1}{n-1}}}{\delta}(\mathrm{diam}(\Omega))^{2} for \Omega\subset\mathbb{R}^{n} , \delta\in(0,n^{\frac{n}{n-1}}\omega_{n}^{\frac{1}{n-1}}) , where \omega_{n} denotes the volume of the unit ball in \mathbb{R}^{n} , and u vanishes on the boundary with -\operatorname{div}(|\nabla u|^{n-2}\nabla u)=f\in L^{1}(\Omega) . An inequality of similar form for the complex Monge-Ampère operator were obtained by [ 7 , 2 , 3 ] . In this paper, we study the analogous theory of k -Hessian equations. Let \Phi^{k}_{0}(\Omega) be the subset of functions in \Phi^{k}(\Omega) with vanishing boundary value. When \partial\Omega is (k-1) -convex, i.e., there exists a constant c_{0}>0 such that \sum_{1\leq i_{1}<\dots<i_{j}\leq n-1}\kappa_{i_{1}}(x)\cdots\kappa_{i_{j}}(x)\geq c_{0} for all x\in\partial\Omega and 1\leq j\leq k-1 , where \kappa_{1}(x),\dots,\kappa_{n-1}(x) are the principal curvatures of \partial\Omega at x , we have \Phi^{k}_{0}(\Omega)\neq\emptyset . We define the k -Hessian integral I_{k}(u)=\int_{\Omega}(-u)S_{k}[u]\,dx and the k -Hessian mass \mathcal{M}_{k}(u)=\int_{\Omega}S_{k}[u]\,dx. Then ||u||_{\Phi^{k}_{0}(\Omega)}:=I_{k}(u)^{\frac{1}{k+1}} is a norm on \Phi^{k}_{0}(\Omega) . As we mentioned above, the k -Hessian measure can be defined for general k -convex functions. We denote by \mathcal{F}^{k}(\Omega) the set of all k -convex functions u such that \mathcal{M}_{k}(u)=\mu_{k}[u](\Omega)<+\infty and \displaystyle\lim_{y\to x}u(y)=0 for any x\in\partial\Omega . The first result of this paper is the following Brezis-Merle type inequality for the k -Hessian equations. Theorem 1.1 . Let u\in\mathcal{F}^{k}(\Omega) . (1) If 1\leq k<n/2 , then for 1\leq p\leq\frac{kn}{n-2k} , there exists C=C(n,k,p,|\Omega|) such that ||u||_{L^{p}(\Omega)}\leq C\mathcal{M}_{k}(u)^{1/k}. (2) If k=\frac{n}{2} , then for 0<\lambda<\alpha_{0} and 1\leq\beta\leq\beta_{0} , there exists C=C(n) such that the following Brezis–Merle type inequality holds: (1.2) \displaystyle\sup\left\{\int_{\Omega}\exp\left(\lambda\left(\frac{-u}{\mathcal{M}_{k}(u)^{\frac{1}{k}}}\right)^{\frac{k\beta}{k+1}}\right)\,dx:u\in\mathcal{F}^{k}(\Omega),\,0<||u||_{\Phi_{0}^{k}(\Omega)}<\infty\right\}\leq\frac{\alpha_{0}C(n)}{\alpha_{0}-\lambda}|\Omega|, where (1.3) \alpha_{0}=n\left[\omega_{n}\binom{n}{k}\right]^{2/n},\quad\beta_{0}=\frac{n+2}{n}. With the Brezis-Merle type inequality in the intermediate case( k=\frac{n}{2} ), we can improve the uniform estimate and get an ABP type estimate. Theorem 1.2 . Let \Omega be a bounded (k-1) -convex smooth domain, k=n/2 and u\in C^{2}(\Omega)\cap C^{0}(\overline{\Omega}) . Let \Phi:\mathbb{R}\to\mathbb{R}^{+} be a positive increasing function with \int_{0}^{\infty}\Phi^{-1/k}(t)\,dt\leq\Lambda<\infty. Then there exist constants c_{1}=c_{1}(n,|\Omega|),c_{2}=c_{2}(n,\Lambda)>0 such that (1.4) \displaystyle\sup_{\Omega}u\leq\sup_{\partial\Omega}u+c_{1}+c_{2}\left(\int_{\Gamma^{-}}S_{k}[-u]\Phi(\log(S_{k}[-u]))\,dx\right)^{1/k}, where \Gamma^{-}=\{x\in\Omega\mid u(x)>0\text{ and }-S_{k}[u](x)>0\} . We can also apply the Brezis-Merle type inequality to study the concentration-compactness principle for the intermediate Hessian equation. Let \{V_{i}\}_{i=1}^{+\infty} be a sequence of nonnegative functions and \{u_{i}\}_{i=1}^{+\infty} be a sequence of weak solutions of S_{k}[u_{i}]=V_{i}e^{-u_{i}} in a domain \Omega with ||V_{i}||_{L^{p}(\Omega)}\leq C_{1},\ ||e^{-u_{i}}||_{L^{p^{\prime}}(\Omega)}\leq C_{2} for some C_{1},C_{2}>0 and some 1<p\leq\infty . Here p^{\prime} is the conjugate exponent of p . Define the blow-up set S=\{x\in\Omega\ |\ \text{there exists a sequence}\;\{x_{i}\}\in\Omega\;\text{such that }x_{i}\to x\;\text{and }u_{i}(x_{i})\to-\infty\}. Then we have Theorem 1.3 . There exists a subsequence \{u_{i_{j}}\} satisfying the following alternative: either (i) \{u_{i_{j}}\} is bounded in L^{\infty}_{loc}(\Omega) or (ii) \{u_{i_{j}}\}\to+\infty uniformly on every compact subset of \Omega or (iii) the blow-up set S of {u_{i_{j}}} is finite and nonempty, and u_{i_{j}}\to+\infty uniformly on compact subsets of \Omega\setminus S . Moreover, in case (iii), the measures V_{i_{j}}e^{-u_{i_{j}}} converge in the sense of measure on \Omega to \mu=\sum_{i}\alpha_{i}\delta_{a_{i}} , where S=\bigcup_{i}{\alpha_{i}} and each coefficient satisfies \alpha_{i}\geq\left(\frac{\alpha_{0}}{p^{\prime}}\right)^{k} . The rest of the paper is organized as follows. In Section 2 , we present the proof of Brezis-Merle type inequality(Theorem 1.1 ). Section 3 is devoted to a Alexandrov-Bakelman-Pucci type estimate for the k -Hessian equation. Finally, in the last section, we apply the Brezis-Merle type inequality to study the concentration-compactness principle. 2. The Brezis-Merle type inequality We make use of the isocapacitary inequalities for the k -Hessian equations to establish the Brezis-Merle type inequalities. Recall the capacity for the k -Hessian equation [ 15 , 11 ] \operatorname{Cap}_{k}(K,\Omega)=\sup\left\{\int_{K}S_{k}[u]\,dx:u\;\text{is k-convex},-1<u<0\right\}. A non-negative Borel measure \mu on \Omega is said to be continuous with respect to the k -Hessian capacity if for every \varepsilon>0 , there exists \delta>0 such that for any open set E\subset\Omega with \operatorname{Cap}_{k}(E,\Omega)<\delta , we have \mu(E)<\varepsilon . It follows from [ 15 ] that k -Hessian measure \mu_{k}[u] is continuous with respect to capacity if either u is bounded in \Omega or there exist constants \varepsilon>0 and C>0 such that \mu_{k}[u](B_{r}(x))\leq Cr^{n-2k+\varepsilon} whenever B_{r}(x)\subset\Omega . We will use the following isocapacitary inequalities. Theorem 2.1 . [ 19 , Theorem 1.2] (i) If 1\leq k<\frac{n}{2} and 1\leq q\leq\frac{n(k+1)}{n-2k} , there exists a constant C=C(n,k,q,|\Omega|)>0 such that for any E\subset\Omega , |E|\leq C\cdot\operatorname{Cap}_{k}(E,\Omega)^{\frac{q}{k+1}}. (ii) If k=\frac{n}{2} and 1\leq\beta\leq 1+\frac{1}{k} , there exists a constant C(n)>0 such that for any E\subset\Omega , |E|\cdot\exp\left(\alpha_{0}\frac{1}{\operatorname{Cap}_{k}(E,\Omega)^{\frac{\beta}{k+1}}}\right)\leq C(n)|\Omega|, where \alpha_{0} is given in ( 1.3 ). In order to establish the Brezis-Merle inequality for general k -convex functions, we need the following comparison principle. Theorem 2.2 . [ 15 , Theorem 3.3] Let \Omega\subset\mathbb{R}^{n} be a bounded domain and let u,v be k -convex functions such that u=v=\phi continuously on \partial\Omega . If the k -Hessian measures \mu_{k}[u] and \mu_{k}[v] are both continuous with respect to the k -Hessian capacity, then \mu_{k}[u]\bigl(\{u<v\}\bigr)\geq\mu_{k}[v]\bigl(\{u<v\}\bigr). Remark 2.3 . The theorem also holds if the condition “ u=v=\phi continuously on \partial\Omega ” is replaced by u\geq v on \partial\Omega . Proof of Theorem 1.1 . Define K_{t}:=\{x\in\Omega:u(x)<-t\} , t>0 . Let u_{K_{t}} be the extremal function with respect to \operatorname{Cap}_{k}(K_{t},\Omega) and let u_{j}\in\Phi_{0}^{k}(\Omega) be a sequence converging to u