Infinite transitivity and polynomial vector fields

Rafael B. Andrist, Ivan Arzhantsev · 2026

Read Infinite transitivity and polynomial vector fields by Rafael B. Andrist, Ivan Arzhantsev (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2605-20480.txt.

Introduction

Rafael Andrist and Ivan Arzhantsev prove a sweeping reach result for symmetry groups of the two-variable complex plane. They take two root subgroups, groups that shift one coordinate by a polynomial in the other. The combined group's diagonal action can then sweep almost every point of any finite power of the plane. The proof rests on the Lie algebra of two-variable polynomials under the standard Poisson bracket.

Catalog record

CREATORRafael B. Andrist, Ivan Arzhantsev
YEAR2026
COLLECTIONPAPERS
FORMATTXT
FILE SIZE38247
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2605.20480

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

Infinite transitivity and polynomial vector fields Rafael B. Andrist, Ivan Arzhantsev Abstract . We prove that for many pairs H 1 , H 2 H_{1},H_{2} of root subgroups of the automorphism group Aut ⁡ ( ℂ 2 ) \Aut({\mathbb{C}}^{2}) the diagonal action of the group generated by H 1 , H 2 H_{1},H_{2} on ( ℂ 2 ) m ({\mathbb{C}}^{2})^{m} has an open orbit for any positive integer m m . The result is based on the study of the Lie algebra of polynomials in two variables with the standard Poisson bracket. 2020 Mathematics Subject Classification Primary 14R10, 17B66; Secondary 14R20, 22E65, 32M17 1. Introduction In this work, we study the properties of multiple transitivity for groups acting on the affine plane {\mathbb{C}}^{2} . It follows from the classical Jung theorem that the group of polynomial automorphisms \Aut({\mathbb{C}}^{2}) is generated by the two-dimensional torus of diagonal matrices and the one-parameter unipotent subgroups of the form L_{r}\colon(x+\alpha y^{r},y),\quad\alpha\in{\mathbb{C}}\quad\text{and}\quad R_{s}\colon(x,y+\beta x^{s}),\quad\beta\in{\mathbb{C}}, where r and s run through the non-negative integers. Let us consider the group G_{r,s} generated by the two subgroups L_{r} and R_{s} . We recall that an action of a group G on a set X is infinitely transitive if for any positive integer m and any two m -tuples of pairwise distinct points in X there is an element g\in G that sends the first tuple to the second one. It is easy to see that the group G_{r,s} acts on {\mathbb{C}}^{2} with an open orbit. By [ 7 ] , the action of G_{r,s} on its open orbit is infinitely transitive only if rs=2 . It is asked in [ 7 ] whether rs=2 implies that the action is indeed infinitely transitive. The positive answer is given in [ 13 , 9 ] . We give another proof of this result in Section 4 . This proof connects the property of infinite transitivity of an action with the structure of a tangent algebra, and therefore appears promising for further generalizations. Let us say that an action of a group G on an algebraic variety X is generically infinitely transitive if for any positive integer m the diagonal action of G on X^{m}:=X\times\ldots\times X ( m times) has a Zariski open orbit. Clearly, any infinitely transitive action is generically infinitely transitive. The main result of this paper (Theorem 18 ) claims that the action of G_{r,s} on {\mathbb{C}}^{2} is generically infinitely transitive if and only if rs\geqslant 2 . It is easy to see that the condition rs\leqslant 1 is equivalent to the fact that G_{r,s} is a linear algebraic group. The proof of Theorem 18 is based on the study of subalgebras of the Lie algebra of polynomials {\mathbb{C}}[x,y] with the standard Poisson bracket; see Section 2 . We show that for the subalgebra \lie(x^{p},y^{q}) generated by the two monomials x^{p} and y^{q} we have \lie(x^{p},y^{q})\cap{\mathbb{C}}[x]=x^{p}{\mathbb{C}}[x^{pq-p-q}]\quad\text{and}\quad\lie(x^{p},y^{q})\cap{\mathbb{C}}[y]=y^{q}{\mathbb{C}}[y^{pq-p-q}], provided \min(p,q)\geq 2 and \max(p,q)\geq 3 ; see Proposition 7 . Moreover, the subalgebra \lie(x^{p},y^{q}) has finite codimension in {\mathbb{C}}[x,y] if and only if (p,q)=(2,3) or (3,2) . In the same Section 2 we find explicit pairs of derivations that generate the Lie algebra of polynomial vector fields on {\mathbb{C}}^{2} with zero divergence or with constant divergence; see Propositions 1 , 11 and Theorem 10 . These results complement the results of [ 1 , 2 , 3 ] , where explicit finite generating sets for many Lie algebras of polynomial vector fields are found. In order to go from Lie algebras to the corresponding transformation groups, we use Proposition 23 on approximation of polynomial flows. This result goes back to Varolin [ 14 ] , some of its versions were used in [ 1 , 2 , 3 ] . This approach is applicable not only to the affine plane {\mathbb{C}}^{2} , but also to other affine surfaces; see Proposition 25 for the case of Danielewski surfaces. Denote by {\mathbb{G}}_{a} the additive group of the ground field {\mathbb{C}} . Any non-trivial regular action {\mathbb{G}}_{a}\times X\to X on an algebraic variety X gives rise to a subgroup H in \Aut(X) isomorphic to {\mathbb{G}}_{a} . Such subgroups are called {\mathbb{G}}_{a} -subgroups of \Aut(X) . The subgroups L_{r} and R_{s} are examples of {\mathbb{G}}_{a} -subgroups in \Aut({\mathbb{C}}^{2}) . These are precisely the {\mathbb{G}}_{a} -subgroups normalized by the two-dimensional torus of diagonal matrices. Such {\mathbb{G}}_{a} -subgroups are called root subgroups. Some results on transitivity properties for groups generated by {\mathbb{G}}_{a} -subgroups can be found in [ 5 , 6 , 7 ] and references therein. In Corollary 29 , we use the fact that the Lie algebra of polynomial vector fields with zero divergence on {\mathbb{C}}^{2} is 2-generated and construct {\mathbb{G}}_{a} -subgroups H_{1},H_{2} (one them is not root) such that the group G generated by H_{1},H_{2} is infinitely transitive on {\mathbb{C}}^{2} . It is important to notice that the results on Lie algebras in Section 2 are obtained by algebraic methods and are valid over any field of characteristic zero. At the same time, the results on transitivity properties of group actions are based on Propositions 23 , 27 , the proofs of which use analytical methods, in particular the Implicit Function Theorem. It would be interesting to obtain criteria for infinite transitivity or generic infinite transitivity in terms of the corresponding Lie algebras applying purely algebraic arguments. 2. Some results on Lie algebras Let {\mathbb{K}} be a field of characteristic zero. Consider the Lie algebra L of all polynomial vector fields f_{1}(x,y)\frac{\partial}{\partial x}+f_{2}(x,y)\frac{\partial}{\partial y} on {\mathbb{K}}^{2} with zero divergence, i.e., \frac{\partial f_{1}(x,y)}{\partial x}+\frac{\partial f_{2}(x,y)}{\partial y}=0. well known that L coincides with the Lie algebra of all volume-preserving vector fields, all symplectic vector fields, and all Hamiltonian vector fields on {\mathbb{K}}^{2} . In particular, vector fields from L are precisely vector fields of the form V_{f}:=\frac{\partial f(x,y)}{\partial y}\frac{\partial}{\partial x}-\frac{\partial f(x,y)}{\partial x}\frac{\partial}{\partial y} for some f\in{\mathbb{K}}[x,y] . As for the commutator, we have [V_{f},V_{h}]=V_{\{f,h\}},\quad\text{where}\quad\{f,h\}=\frac{\partial f(x,y)}{\partial x}\frac{\partial h(x,y)}{\partial y}-\frac{\partial f(x,y)}{\partial y}\frac{\partial h(x,y)}{\partial x} is the standard Poisson bracket on {\mathbb{K}}[x,y] . We denote by A the Lie algebra ({\mathbb{K}}[x,y],\{\cdot,\cdot\}) . The map f\mapsto V_{f} establishes an isomorphism of Lie algebras between A/I and L , where I is the central ideal of constants in A . As the first result, we show that the Lie algebra L is 2-generated. We provide a short direct proof. Proposition 1 . The Lie algebra L is generated by two locally nilpotent derivations \partial_{1}=y^{2}\frac{\partial}{\partial x}\quad\text{and}\quad\partial_{2}=\frac{\partial}{\partial x}-x\frac{\partial}{\partial y}. Proof. Notice that \partial_{1}=\frac{1}{3}V_{y^{3}} and \partial_{2}=\frac{1}{2}V_{x^{2}+2y} . So it suffices to show that the polynomials x^{2}+2y and y^{3} generate the Lie algebra A . Let A_{0} be the Lie subalgebra in A generated by x^{2}+2y and y^{3} . We have \{x^{2}+2y,y^{3}\}=6xy^{2},\ \{x^{2}+2y,xy^{2}\}=4x^{2}y-2y^{2},\ \{x^{2}+2y,2x^{2}y-y^{2}\}=4x^{3}-12xy, \{x^{2}+2y,x^{3}-4xy\}=6y-12x^{2},\ \ \{x^{2}+2y,y-2x^{2}\}=10x,\ \ \{x^{2}+2y,x\}=-2. Since the polynomials x^{2}+2y and y-2x^{2} are not proportional, we conclude that 1,x,x^{2},y lie in A_{0} . Further, we obtain \{\{x^{2},y^{l}\},y^{l}\}=2l^{2}y^{2(l-1)}. This implies that if y^{l}\in A_{0} then y^{2(l-1)}\in A_{0} and 2(l-1)>l for l\geqslant 3 . Since we have \{x,y^{l}\}=ly^{l-1}, we conclude that y^{s}\in A_{0} for all s\in{\mathbb{Z}}_{\geqslant 0} . In the same way we have \{\{\{y^{3},x^{k}\},x^{k}\},x^{k}\}=6k^{3}x^{3(k-1)}. This shows that if x^{k}\in A_{0} then y^{3(k-1)}\in A_{0} and 3(k-1)>k for k\geqslant 2 . Since we have \{x^{k},y\}=kx^{k-1}, we conclude that x^{m}\in A_{0} for all m\in{\mathbb{Z}}_{\geqslant 0} .

Read the complete text