Stochastic homogenization of a nonconvex Hamilton-Jacobi equation (Q745555): Difference between revisions

This paper deals with the Hamilton-Jacobi equation \(u^{\varepsilon}_{t}+\left(| D u^{\varepsilon}|^2-1 \right)^2-V\left({x\over \varepsilon}\right) =0\) in \(\mathbb R^d\times(0,\infty)\), \(d\geq1\). The potential \(V\) is assumed to be a bounded, stationary-ergodic random potential. Let \(\Omega=\mathrm{BUC}(\mathbb R^d)\) be the space of real-valued, bounded and uniformly continuous functions on \(\mathbb R^d\); \({\mathcal F}\) is the \(\sigma\)-algebra on \(\Omega\) generated by point-wise evaluations; the translation group \(T_{y}: \Omega\to\Omega\) is defined by \((T_{y}V)(x)=V(x+y)\). It is considered a probability measure \(P\) on \((\Omega,{\mathcal F})\) satisfying the following properties: there exists \(K_0>0\) such that \(P\left[\displaystyle\sup_{x\in \mathbb R^d}| V(x)|\leq K_0\right]=1\) (uniform boundedness), for every \(E\in{\mathcal F}\) and \(y\in \mathbb R^d\), \(P[E]=P[T_{y}E]\)(stationarity) and \(P[\bigcap_{z\in \mathbb R^d}T_{z}E]\in\{0,1\}\) (ergodicity). Let \(u^{\varepsilon}(\cdot,g)\in C(\mathbb R^d\times[0,\infty))\) be a unique solution of considere Hamilton-Jacobi equation subject to the initial condition \(u^{\varepsilon}(x,0,g)=g(x)\). All differential equations in this paper are to be interpreted in the viscosity sense. The main result is following. There exists \(\bar H\in C(\mathbb R^d)\) satisfying \(\bar H(p)\to+\infty\) as \(| p|\to\infty\) such that, if \(u(x,t,g)\) is the unique solution of the equation \(u_{t}+\bar H(Du)=0\) in \(\mathbb R^d\times(0,\infty)\) subject to the initial condition \(u(x,0,g)=g(x)\),\(\forall g\in\mathrm{BUC}(\mathbb R^d)\), then \(P\left[\forall g\in\mathrm{BUC}(\mathbb R^d),\forall k>0,\displaystyle\limsup_{\varepsilon\to0}\sup_{(x,t)\in B_{k}\times[0,k]}| u^{\varepsilon}(x,t,g)-u(x,t,g)|=0\right]=1\).