Distinguishing filling invariants associated to conjugacy in groups

Conan Gillis and Timothy Riley · 2025

Read Distinguishing filling invariants associated to conjugacy in groups by Conan Gillis and Timothy Riley (2025) 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-2506-19264.txt.

Introduction

Three cost gauges for the conjugacy puzzle in groups turn out to be truly independent, and Conan Gillis and Timothy Riley prove it with fresh examples, in work from twenty twenty-five. The gauges grew from a nineteen ninety-eight frame that tied word-merging costs to area and length costs. Kinship among the three was long assumed in part.

Their example groups split every pair: no two gauges fix the third. Each gauge thus earns its keep as its own measure of group hardness. The result closes a basic question in geometric group theory and warns against shortcut claims that one cost can stand for all.

Catalog record

CREATORConan Gillis and Timothy Riley
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE92 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2506-19264

Browse the full library · About the Tower of Babel

Full text

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

Distinguishing filling invariants associated to conjugacy in groups Thanks: The authors gratefully acknowledge the support of NSF Grants DGE–2139899 (Gillis) and OIA-2428489 (Riley). Conan Gillis and Timothy Riley Abstract . Brick and Corson introduced annular Dehn functions in 1998 to quantify the conjugacy problem for finitely generated groups and gave the fundamental relationships between it, the Dehn function, and the conjugator length function. We furnish the theory with diverse examples groups. In particular, we show that these three invariants are independent—no two of the three functions determine the other. 2020 Mathematics Subject Classification: 20F65, 20F10, 20F06 Key words and phrases: conjugacy problem, Conjugator length, Dehn function 1. Introduction When a word u represents 1 in a finitely presented group G , it admits a disk-diagram (a van Kampen diagram ) witnessing to how this follows from the defining relations. When words u and v represent conjugate elements, the corresponding witness is an annular diagram . The annular Dehn function \hbox{\rm Ann}:\mathbb{N}\to\mathbb{N} of a finitely presented group G was introduced by Brick and Corson in 1998 [ BC98 ] . They defined \hbox{\rm Ann}(n) to be the minimal N such that whenever u and v are words that represent conjugate elements of G and whose lengths sum to at most n , there is an annular diagram with at most N faces such the inner boundary is labelled by u and the outer boundary by v . As Brick and Corson observed, the annular Dehn function is closely related to two other filling functions. One is the Dehn function \hbox{\rm Area}:\mathbb{N}\to\mathbb{N} of G , which is defined so that \hbox{\rm Area}(n) is the minimal N such that whenever a word u of length at most n represents 1 in G , it admits a van Kampen diagram with at most N faces. The other is the conjugator length function \hbox{\rm CL}:\mathbb{N}\to\mathbb{N} of G, which is defined so that \hbox{\rm CL}(n) is the minimal N such that whenever u and v are words that represent conjugate elements of G and whose lengths sum to at most n , there is a word w of length at most N such that uw=wv in G . Equivalently, there is an annular diagram for the pair u and v for which there is a path in the 1 -skeleton of length at most N from the start of u on one of the two boundary components to the start of v on the other. Section 2 contains more detailed definitions. Dehn functions have been studied extensively— [ Bri02 , BRS07 , Ger93 , Sap11 ] are surveys. Conjugator length functions have also received considerable attention— [ BRS ] is a recent account. Much less is known about annular Dehn functions. Brick and Corson [ BC98 ] observed the fundamental relationships: Theorem 1.1 (Brick and Corson’s inequalities) . For any finitely presented group G , (1) \hbox{\rm Area}(n)\ \leq\ \hbox{\rm Ann}(n)\ \leq\ \hbox{\rm Area}(2\hbox{\rm CL}(n)+n) and (2) \hbox{\rm CL}(n)\ \leq\ \frac{n}{2}+M\cdot\hbox{\rm Ann}(n), where M>0 is a constant depending on the presentation for G . In brief, the second bound in ( 1 ) comes from the observation that for a pair of words u and v that represent conjugate elements of G , and for a minimal length conjugator w , we can form an annular diagram by identifying the two w -labelled sides in a minimal-area van Kampen diagram for w^{-1}uwv^{-1} . The bound in ( 2 ) comes from taking M to be the length of the longest defining relation, since then the right-hand side is an upper bound on the number of edges in an annular diagram. The purpose of this article is to furnish the theory of annular Dehn functions with some examples which show that it harbors some richness and subtlety: specifically, we show that \hbox{\rm Area}(n) , \hbox{\rm CL}(n) , and \hbox{\rm Ann}(n) are independent invariants and ( 1 ) and ( 2 ) need not be sharp. These conclusions will be corollaries of the estimates that are summarized in the following theorem. The conventions are a^{b}=aba^{-1} and [a,b]=aba^{-1}b^{-1} , and for functions f,g:\mathbb{N}\to\mathbb{N} , f\lesssim g if there exists a constant C>0 such that f(n)\leq Cg(Cn+n)+Cn+C for all n . We write f\simeq g when f\lesssim g and g\lesssim f . Up to \simeq , each of \hbox{\rm Ann}(n) , \hbox{\rm CL}(n) , and \hbox{\rm Area}(n) do not depend on the choice of finite presentation for G . Theorem 1.2 . Up to \simeq , the following finitely presented groups have the following Dehn functions, conjugator length functions, and annular Dehn functions (for all d,m\geq 1 ): \hbox{\rm Area}(n) \hbox{\rm Ann}(n) \hbox{\rm CL}(n) G_{1}=\mathcal{H}_{3}(\mathbb{Z}) n^{3} n^{4} n^{2} G_{2}=\mathrm{BS}(1,2) 2^{n} 2^{n} n G_{3}=G_{1}\times G_{2} 2^{n} 2^{n} n^{2} G_{4} 2^{n} 2^{n^{2}} n^{2} G_{5,d}=\mathbb{Z}^{d}\rtimes\mathbb{Z} filiform n^{d+1} n^{d} G_{6,m} n^{3} n^{m+1} G_{7}=G_{5,3}\ast G_{6,20} n^{4} \left.\rule{0.0pt}{11.38109pt}\right\}\simeq n^{21} G_{8}=G_{5,4}\ast G_{6,20} n^{5} n^{21} where \begin{array}[]{ll}G_{1}&\!\!\!=\ \langle a,b,c\mid[a,c],[b,c],[a,b]c^{-1}\rangle\\ G_{2}&\!\!\!=\ \langle a,s\mid s^{a}s^{-2}\rangle\\ G_{4}&\!\!\!=\ \langle a,b,c,d,s\mid[a,b]c^{-1},[a,c],[b,c],[b,d],s^{a}s^{-2},s^{d}s^{-2}\rangle\\ G_{5,d}&\!\!\!=\ \langle a_{1},\ldots,a_{d},t\mid a_{i}^{t}=a_{i}a_{i-1}\ \ \forall i>1,\ [a_{1},t]=1,\ [a_{i},a_{j}]=1\ \ \forall i\neq j\rangle\end{array} and G_{6,m} is per Definition 1.3 below. To be clear, the assertions made in the table above include that \hbox{\rm Ann}_{G_{7}}(n)\simeq\hbox{\rm Ann}_{G_{8}}(n) , but we do not determine their growth rates. We also do not determine \hbox{\rm Ann}_{G_{5,d}}(n) or \hbox{\rm Ann}_{G_{6,m}}(n) . The Dehn function estimates for G_{1} and G_{2} in this theorem are well-known. Those for G_{5,d} and G_{6,m} are from [ BP94 , GHR03 ] . The conjugator length estimates for G_{1} , G_{2} , G_{5,d} , and G_{6,m} are from [ BRS , Sal16 , BRa , BRb ] . (We discuss these references in more detail in Section 3 .) The remaining estimates are established in this paper. The construction of G_{4} is also novel. Definition 1.3 . The groups G_{6,m} of [ BRb ] are, for all m\geq 1 , the central extensions of \mathbb{Z}^{m+2}=\langle a_{1},\ldots,a_{m},b_{1},b_{2}\rangle by \mathbb{Z}^{m}=\langle c_{1},\ldots,c_{m}\rangle defined by the relations b_{1}a_{i}=a_{i}b_{1}c_{i} for i=1,\ldots,m , b_{2}a_{i}=a_{i}b_{2}c_{i+1}^{-1} for i=1,\ldots,m-1 , and commutation relations for the pair of generators (b_{2},a_{m}) , and all pairs of generators (a_{i},a_{j}) , (b_{i},b_{j}) , (c_{i},c_{j}) , (b_{i},c_{j}) , and (a_{i},c_{j}) . Corollary 1.4 . The Dehn function, the conjugator length function, and the annular Dehn function are independent invariants for finitely presented groups. That is, for any two of these invariants, there is a pair of finitely presented groups for which those two invariants agree, but the third differs (all up to \simeq ). Proof. By Theorem 1.2 , the groups G_{7} and G_{8} have different Dehn functions, but the other two functions the same. Likewise, compare G_{2} and G_{3} for conjugator length function, and compare G_{3} and G_{4} for annular Dehn function. ∎ Corollary 1.5 . There exist finitely presented groups for which the inequalities \hbox{\rm Area}(n)\ \stackrel{{\scriptstyle\textup{i}}}{{\lesssim}}\ \hbox{\rm Ann}(n)\ \stackrel{{\scriptstyle\textup{ii}}}{{\lesssim}}\ \hbox{\rm Area}(n+\hbox{\rm CL}(n)) of Brick and Corson are (1) both sharp, (2) both not sharp, (3) i is sharp but ii is not, and (4) ii is sharp but i is not, where by ‘sharp’ we mean that \preceq can be replaced \simeq . Proof. Our examples are: (1) G_{2} , (2) G_{1} , (3) G_{3} , and (4) G_{4} . ∎ An outine of this article. Section 2 provides preliminaries concerning van Kampen diagrams, annular

Read the complete text