English

An eigenvalue problem for fully nonlinear elliptic equations with gradient constraints

Analysis of PDEs 2015-09-01 v3

Abstract

We consider the problem of finding λR\lambda\in \mathbb{R} and a function u:RnRu:\mathbb{R}^n\rightarrow\mathbb{R} that satisfy the PDE max{λ+F(D2u)f(x),H(Du)}=0,xRn. \max\left\{\lambda + F(D^2u) -f(x),H(Du)\right\}=0, \quad x\in \mathbb{R}^n. Here FF is elliptic, positively homogeneous and superadditive, ff is convex and superlinear, and HH is typically assumed to be convex. Examples of this type of PDE arise in the theory of singular ergodic control. We show that there is a unique λ\lambda^* for which the above equation has a solution uu with appropriate growth as x|x|\rightarrow \infty. Moreover, associated to λ\lambda^* is a convex solution uu^* that has bounded second derivatives, provided FF is uniformly elliptic and HH is uniformly convex. It is unknown whether or not uu^* is unique up to an additive constant; however, we verify this is the case when n=1n=1 or when F,f,HF, f,H are "rotational."

Keywords

Cite

@article{arxiv.1412.8011,
  title  = {An eigenvalue problem for fully nonlinear elliptic equations with gradient constraints},
  author = {Ryan Hynd},
  journal= {arXiv preprint arXiv:1412.8011},
  year   = {2015}
}
R2 v1 2026-06-22T07:44:32.579Z