Interior regularity of solutions to a complex Monge-Ampère equation (Q1774805)

A domain \(\Omega \subset {\mathbb C}^n\) is called hyperconvex if for every \(z_0 \in \partial \Omega\), there exists \(\rho \in {\mathcal PSH} (\Omega)\) such that \(\rho < 0\) in \(\Omega\) and \(\lim_{ z \to z_0} \rho (z) =0\). According to arguments of \textit{Z. Błocki} [Ann. Sc. Norm. Super. Pisa, Cl. Sci., IV. Ser. 23, No. 4, 721--747 (1996; Zbl 0878.31003)], one may in addition assume that \(\rho\) is \({\mathcal C}^\infty\)-smooth, that the limit of \(\rho\) is zero on \(\partial \Omega\) and that \(\det ({\partial^2 \rho \over \partial z_j \partial \bar z_k}) \geq 1\). The domain is said to satisfy the \textsl{non-precipitiousness condition} if, moreover, there exists such a function \(\rho\) whose real gradient is bounded in \(\Omega\): \[ \sup \left\{ \left| {\partial \rho \over \partial x_j} (z) \right|, \;\left| {\partial \rho \over \partial y_j} (z) \right| : \;z \in \Omega, \;j= 1, \dots, n \right\} < \infty, \] where \(z_j = x_j + iy_j\). The (announced) main result is as follows. Let \(\Omega\) be hyperconvex and let \(f \in {\mathcal C}^\infty (\overline{\Omega} \times {\mathbb R})\) be a strictly positive function, increasing with the second variable. If \(\Omega\) satisfies the non-precipitiousness condition, then the problem \[ \det \left( {\partial^2 u \over \partial z_j \partial \bar z_k} \right) = f(z, u(z)) \quad \text{in } \Omega; \qquad \lim_{z \to z_0} u(z) = 0, \quad \text{for all } z_0 \in \partial \Omega, \] has a unique smooth strictly plurisubharmonic solution \(u\), whose real gradient is moreover bounded in \(\Omega\). This main result strongly relies on recent works of Z.~Błocki and on the existence of the solutions to the Monge-Ampère equation published by \textit{F.~Schulz} two decades ago [J. Reine Angew. Math. 348, 88--93 (1984; Zbl 0521.35010) ]. However, recently, \textit{Z.~Błocki} has observed a gap in the proof of F.~Schulz [Bull. Aust. Math. Soc. 68, No. 1, 81--92 (2003; Zbl 1067.35028)], which seems not to be filled up to now, so that the applications of the paper under review are ground missing.