On the reconstruction of polynomial automorphisms from their face polynomials (Q1196827): Difference between revisions

Let \(k\) be a field, \(F=(F_ 1,\dots,F_ n)\): \(k^ n\to k^ n\) a polynomial automorphism and \(F_ i(X_ j=0):=F_ i(X_ 1,\dots,X_{j-1},0,X_{j+1},\dots,X_ n)\), \(1\leq i,j\leq n\), its face polynomials. It is known [cf. \textit{J. H. McKay} and \textit{S. S.-S. Wang}, J. Pure Appl. Algebra 52, No. 1/2, 103-119 (1988; Zbl 0664.13001)] that \(F\) is completely determined by its face polynomials. In the case \(n=2\) \textit{J. H. McKay} and \textit{S. S.-S. Wang} [J. Pure Appl. Algebra 40, 245- 257 (1986; Zbl 0622.13003)] gave an explicit inversion formula in terms of the resultant of the face polynomials. The authors generalize the inversion formula to higher dimensions under some restricted assumptions on \(F\). In the proof, they use higher dimensional resultants instead of the classical ones. --- Furthermore, they give an algorithm, using Gröbner bases, to reconstruct the inverse of a polynomial automorphism (without any assumptions) from its face polynomials.