Formal desingularization of surfaces: The Jung method revisited (Q2377955): Difference between revisions

Constructive desingularization of varieties (in characteristic zero) is a very important tool in modern algebraic geometry, but unfortunately implementations for the general case (i.e. arbitrary dimension) suffer from the high computational complexity of the construction itself. In the case of curves and surfaces, other approaches were already known, when Hironaka proved his famous result on the existence of resolution of singularities in characteristic zero in the 1960s, but were no longer in the center of interest after the general case was proved. The important exception here is the curve case, where the Newton-Puiseux expansion locally provides the desired data and is successfully applied for many practical purposes. In this article, the author studies desingularization of hypersurfaces in \({\mathbb P}^3\) over a field of characteristic zero on the basis of these considerations, using Jung's (local) approach to resolution of surface singularities. To this end, he first discusses the formal setting in which he is working and develops Jung's construction to the point where he can give explicit (a priori theoretical) algorithms for its various subtasks. To turn these considerations into an implementation, he then passes to a discussion of the practical aspects of multivariate algebraic power series and gives a new proof of the fact that the concept of rational Puiseux parametrizations can be extended to multivariate quasiordinary polynomials. At the end of the article, the author also mentions the performance of his implementation of the described algorithm in MAGMA and its availability, as well as the still open problem to obtain a dual graph of the resolution along these lines.