The Infinity Laplacian eigenvalue problem: reformulation and a numerical scheme
Farid Bozorgnia, Leon Bungert, Daniel Tenbrinck
TL;DR
This study tackles the challenging infinity Laplacian eigenvalue problem by reformulating it into a single, sign-agnostic framework using the operator $H_\Lambda$, and proves its equivalence to the classical $F_\Lambda$ formulation in the viscosity-solution sense. Building on this, the authors develop consistent monotone grid schemes for ground states and higher eigenfunctions, establishing subsequential convergence to viscosity solutions via a Barles–Souganidis-type argument and without relying on a comparison principle. They implement a fixed-point (Euler) iteration to solve the discrete system and demonstrate robust numerical performance on diverse domains, including clear evidence that the infinity ground state generally differs from the infinity harmonic potential on a square. The work advances the numerical study of infinity-Laplacian eigenfunctions, enabling systematic exploration of conjectures and geometric characterizations, while also highlighting open questions such as second eigenvalues on general domains and convergence rates. Overall, the method provides a practical tool for computing infinity-Laplacian eigenfunctions and investigates fundamental qualitative properties of these nonlinear, non-unique objects.
Abstract
In this work, we present an alternative formulation of the higher eigenvalue problem associated to the infinity Laplacian, which opens the door for numerical approximation of eigenfunctions. A rigorous analysis is performed to show the equivalence of the new formulation to the traditional one. Subsequently, we present consistent monotone schemes to approximate infinity ground states and higher eigenfunctions on grids. We prove that our method converges (up to a subsequence) to a viscosity solution of the eigenvalue problem, and perform numerical experiments which investigate theoretical conjectures and compute eigenfunctions on a variety of different domains.
