An algorithmic approach to the Polydegree Conjecture for plane polynomial automorphisms

From MaRDI portal
Publication:2318359



Abstract: We study the interaction between two structures on the group of polynomial automorphisms of the affine plane: its structure as an amalgamated free product and as an infinite-dimensional algebraic variety. We introduce a new conjecture, and show how it implies the Polydegree Conjecture. As the new conjecture is an ideal membership question, this shows that the Polydegree Conjecture is algorithmically decidable. We further describe how this approach provides a unified and shorter method of recovering existing results of Edo and Furter.


By the classical Jung-van der Kulk Theorem the group \({\mathcal G}={\mathcal G}({\mathbb C})\) of the polynomial automorphisms of the affine plane is an amalgamated free product of the affine subgroup \({\mathcal A}({\mathbb C})\) and the triangular subgroup \({\mathcal B}({\mathbb C})\) over their intersection. Hence \(\theta\in{\mathcal G}({\mathbb C})\) has the form \(\theta=\alpha_0\tau_1\alpha_1\cdots\tau_k\alpha_k\), where \(\alpha_i\in{\mathcal A}({\mathbb C})\) and \(\tau_i\in{\mathcal B}({\mathbb C})\). This decomposition is not unique but the degrees \(d_i\) of the triangular automorphisms \(\tau_i\) are invariants of \(\theta\). One defines the polydegree of \(\theta\) as the sequence \((d_1,\ldots,d_k)=(\deg\tau_1,\ldots,\deg\tau_k)\). Let \({\mathcal G}_{(d_1,\ldots,d_k)}\) be the set of all automorphisms of polydegree \((d_1,\ldots,d_k)\) and let \(\overline{{\mathcal G}_{(d_1,\ldots,d_k)}}\) be its closure in the Zariski topology. The Polydegree Conjecture is related with the description of \(\overline{{\mathcal G}_{(d,e)}}\) in the length 2 case when \(d,e>2\). Partial results in this direction were obtained in [\textit{E. Edo}, Acta Math. Vietnam. 32, No. 2--3, 155--168 (2007; Zbl 1147.14030)] and [\textit{J. P. Furter}, J. Lond. Math. Soc., II. Ser. 91, No. 1, 180--202 (2015; Zbl 1341.14024)]. In the paper under review, the authors conjecture more generally that if \(k<l\) and the degree sequences \((d_1,\ldots,d_k)\) and \((e_1,\ldots,e_l)\) satisfy \(d_1+\cdots+d_k<e_1+\cdots+e_l\), then \({\mathcal G}_{(d_1,\ldots,d_k)}\subset\overline{{\mathcal G}_{(e_1,\ldots,e_l)}}\). The authors take a direct approach with the hopes that the techniques will prove more generalizable to the length 3 case. They establish three successively stronger results which provide means of concluding that their conjecture holds in certain cases. The authors show that their methods give a unified and shorter method of recovering the results of Edo and Furter. They also prove that the Polydegree Conjecture is algorithmically decidable for any fixed \(d\) and \(e\). Using a computer, the authors show that the conjecture holds for \(e=3\) and \(d<50\) and for \(e=4\) and \(d<20\). Finally, they affirmatively answer a question of Arzhantsev regarding the infinite transitivity of certain group actions on \({\mathbb C}^2\).





Describes a project that uses

Uses Software






This page was built for publication: An algorithmic approach to the Polydegree Conjecture for plane polynomial automorphisms

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2318359)