1

Consider the infinity harmonic functions, i.e. solutions of the equation

$$ \Delta_\infty u = \langle Du, D^2 u \, Du \rangle = 0. $$

It is known that the solutions are everywhere differentiable (continuous differentiability is an open question), and in the plane it is known that the solutions are $C^{1,\alpha}$.

My question is: Is anything known about the second derivative of the solutions, or even its trace, the laplacian? Do we have, for example, that $D^2u \in L^1$, or even Radon measure? Or do we have that $\Delta u$ is a Radon measure?

These properties do hold for the Aronsson solution $$u(x,y) = x^{4/3}-y^{4/3}$$ which is taken as the archetypal solution to infinity Laplace equation.

(I asked the question earlier on Mathoverflow, https://mathoverflow.net/q/162046/1445, and got no attention whatsoever.)

Tommi
  • 1,425
  • For really technical questions like this, if you don't get an answer on MO, you probably won't get one here either. Off the top of my head, I don't see any reason why the Laplacian (which is no longer connected to the structure of your operator) should exhibit any cancellations that makes it behave obviously better than the Hessian. – Willie Wong May 15 '14 at 12:06
  • What do you mean by $D^2u$? The aronssons exaple is not twice diferentiable at the origin. – user29999 Apr 16 '23 at 13:50

0 Answers0