The generalized amalgamated product structure of the tame automorphism group in dimension three (Q2343158)

Consider the automorphism group \(\Aut(K[x_1,\dots,x_n])\) of a polynomial ring \({\mathbb F}[x_1,\dots,x_n]\). It has a \textit{tame automorphism group} as a subgroup, generated by \textit{linear automorphisms} \[ x_i\to \sum_{j=1}^{n}a_{ij}x_j;\;i=1,\dots,n; a_{ij}\in{\mathbb F}\;\det(a_{ij})\neq 0 \] and \textit{unitriangular automorphisms} \[ x_i\to a_ix_i+P_i(x_{i+1},\dots,x_{n}); a_i\in{\mathbb F},\;i=1,\dots,n, \] \(P_i\) are some polynomials. If automorphism is not tame, it called to be \textit{wild}. It is well known that all automorphisms are tame if \(n=2\), and there exist wild automorphisms when \(n=3\) and \(\mathrm{Char}({\mathbb F})=0\), for example Nagata automorphism is wild. (The last fact was established in the famous papers of \textit{I. P. Shestakov} and \textit{U. U. Umirbaev} [J. Am. Math. Soc. 17, No. 1, 197--227 (2004; Zbl 1056.14085)]. The existence of wild automorphisms for free associative algebra (i.e., solution of \textit{Anick conjecture}) case was established by \textit{U. U. Umirbaev} [J. Reine Angew. Math. 605, 165--178 (2007; Zbl 1126.16021)]. Nothing is known for \(n>3\). The proof was based on description of all defining relations of tame automorphism group of \(A^3_K\) [\textit{U. U. Umirbaev}, J. Algebra 314, No. 1, 209--225 (2007; Zbl 1132.17002)]. The goal of the paper is to represent Umirbaev's result of describing defining relations for automorphism groups of free algebras on the language of generalized amalgamated products, introduced by D. Wright, in the spirit of category theory. For \(K\) a field of characteristic zero, it is shown that the tame subgroup \(\mathrm{TA}_3(K)\) of the group \(\mathrm{GA}_3(K)\) of polynomial automorphisms of \(A^3_K\) can be realized as a generalized amalgamated product, specifically, the product of three subgroups, amalgamated along pairwise intersections, in a manner that generalizes the well-known amalgamated free product structure of \(\mathrm{TA}_2(K)\) (which coincides with \(\mathrm{GA}_2(K)\) by Jung's Theorem). The result follows from defining relations for \(\mathrm{TA}_3(K)\) given by U. U. Umirbaev. The proofs based on some nice combinatorial calculations.