Around the Abhyankar-Sathaye conjecture (Q273879): Difference between revisions

This is very nice paper, devoted to the classical Abhyankar-Sathaye conjecture: Conjecture. \(f \in k[x_1,\dots,x_n]\) is an irreducible element whose zero locus in \(\mathbb A^n\) is isomorphic to \(\mathbb A^{n-1}\), then \(f\) is a coordinate. This conjecture is equivalent to the claim that every closed embedding \[ \mathfrak i:\mathbb A^{n-1}\to\mathbb A^n \] is rectifiable, i.e., there is an automorphism \(\sigma\in\mathrm{Aut}(\mathbb A^n)\) such that \(\sigma\circ\mathfrak i:\mathbb A^{n-1}\to\mathbb A^n\) is the standard embedding. This conjecture is linked to investigation of unipotent group action. The main technical result (of own interest) is Theorem 1, based on the fact of triviality of bundles in rational case. Theorem 2 stating partial case of Abhyankar-Sathaye Conjecture for unipotent group action based on this theorem. A rational version of the strengthened form of the Commuting Derivation Conjecture, in which the assumption of commutativity is dropped, is proved. A systematic method of constructing in any dimension greater than 3 the examples answering in the negative a question by \textit{M. El Kahoui} [J. Algebra 289, No. 2, 446--452 (2005; Zbl 1140.13309)] (see Question 1 in the paper) is developed.