Extending knot polynomials of braided Hopf algebras to links
Stavros Garoufalidis, Matthew Harper, Ben-Michael Kohli, Jiebo Song, Guillaume Tahar · 2025
Read Extending knot polynomials of braided Hopf algebras to links by Stavros Garoufalidis, Matthew Harper, Ben-Michael Kohli, Jiebo Song, Guillaume Tahar (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-2505-01398.txt.
Introduction
Stavros Garoufalidis and co-authors spread knot marks from braided algebra forms to links of more than one loop. The work dates to 2025. A rich set of such marks came of late from stiff twist maps of those forms. They were built for knots, one loop each.
The team asks how the marks stretch to links with two or more loops. That stretch lets them match some new marks to known link marks. Those matches back claims made in the first work. The piece ties fresh algebra goods to the old stock of link marks.
Catalog record
| CREATOR | Stavros Garoufalidis, Matthew Harper, Ben-Michael Kohli, Jiebo Song, Guillaume Tahar |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 238 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2505-01398
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Extending knot polynomials of braided Hopf algebras to links Thanks: Key words and phrases: Knots, Links, Nichols algebras, Links–Gould polynomial, ADO invariant, R R -matrices, Yang–Baxter equation. Stavros Garoufalidis, Matthew Harper, Ben-Michael Kohli, Jiebo Song, Guillaume Tahar Abstract . Recently, a plethora of multivariable knot polynomials were introduced by Kashaev and one of the authors, by applying the Reshetikhin–Turaev functor to rigid R R -matrices that come for braided Hopf algebras with automorphisms. We study the extension of these knot invariants to links, and use this to identify some of them with known link invariants, as conjectured in that same recent work. 1. Introduction 1.1. From knot polynomials to link polynomials Suppose one is given a polynomial invariant of knots in 3-space. Is there a natural way to extend it to a polynomial invariant of links? Of course one can extend it by declaring it to vanish for links of more than one component, or that it is the product of the polynomials of each component of the link. Most people would accept, however, that such extensions are not natural, whatever the word “natural” means. This paper aims to give an answer to this question for the polynomial invariants of (long) knots constructed via the Reshetihkin–Turaev functor [ RT ] using as input a rigid R -matrix, as explained in detail in [ Kas1 ] . A rich source of rigid R -matrices was recently discovered by Kashaev and one of the authors in [ GaKa ] to come from braided Hopf algebras (such as Nichols algebras) with automorphisms and their finite dimensional left/right Drinfeld–Yetter modules. This, together with the classification of Nichols algebras gives a systematic way to construct multivariable polynomials of long knots. In the case of rank 1 Nichols algebras, these invariants were identified in [ GaKa ] to be the colored Jones polynomials of a knot [ Jones , RT ] and the ADO polynomial invariants at roots of unity [ ADO ] . In the above reference, 2-variable knot polynomials coming from the simplest rank 2 Nichols algebra were defined, and denoted by \Lambda_{\omega}(t,s) and V_{n}(t,q) , respectively. Our goal is to extend these polynomial invariants of knots to invariants of links. Throughout the paper, all knots and links will be oriented, embedded in S^{3} , and considered up to ambient isotopy. But even more, a rigid R -matrix R\in\mathrm{End}(V\otimes V) on V together with an enhancement h\in\mathrm{End}(V) that satisfies the polynomial equations from [ Ohtsuki , Thm. 3.7] gives invariants of tangles (up to isotopy) that may have closed components. We will review these identities in Section 2.2 below, where we also show that the rigid R -matrices of [ GaKa ] often have a canonically defined enhancement h that satisfies the required identities ( 4a )–( 4d ). Assuming this, a rigid R -matrix on V gives rise to an invariant F of closed links. However, for some rigid R -matrices (e.g., for all discussed in our paper) F is the zero invariant. To obtain a nontrivial invariant, we consider tangles, where a rigid R -matrix gives an \mathrm{End}(V) -valued invariant F_{T} of a (1,1) -tangle T and an \mathrm{End}(V\otimes V) -valued invariant F_{T} of a (2,2) -tangle with upward oriented boundary, by which we mean a tangle with two upward incoming and outgoing arcs and perhaps additional closed components. To get to a scalar-valued invariant of links, we need to ensure: ( P_{1} ) For every (1,1) -tangle T , F_{T} is a scalar multiple of \mathrm{id}_{V} . ( P_{2} ) For every (2,2) -tangle T defining a map F_{T}\in\mathrm{End}(V^{\otimes 2}) with left and right closures T_{1} and T_{2} , we have F_{T_{1}}=F_{T_{2}}\in\mathrm{End}(V) . T T_{1} T T_{2} Figure 1. The left and right closures T_{1} and T_{2} of a (2,2) -tangle T . When ( ( P 1 ) ) is satisfied and T is a (1,1) -tangle, we write F_{T}=\langle F_{T}\rangle\mathrm{id}_{V} . Properties ( ( P 1 ) ) and ( ( P 2 ) ) imply the existence of a scalar invariant of links L obtained by cutting a link along any component to obtain a (1,1) -tangle L^{\mathrm{cut}} . We denote this modified Reshetikhin-Turaev link invariant \text{mRT}_{L}=\langle F_{L^{\mathrm{cut}}}\rangle . Remark 1.1 . When F is determined by a ribbon category, Property ( ( P 1 ) ) is the assumption that the coloring object V is simple. Property ( ( P 2 ) ) is the assumption that V is an ambidextrous object in the sense of [ GPT ] . Once the link invariant is defined, the next question is how to identify it with a known link invariant. For this, observe that any identity of the R -matrix of braids implies a corresponding skein theory for the constructed link invariants. A key example of this exposition is the R -matrix of the Alexander polynomial given in the beautiful exposition of Ohtsuki [ Ohtsuki , Prop. 2.6] , which we review in Section 2.4 below, using a slightly different R -matrix for the purpose of our work. 1.2. Our results With the above preliminaries, we can phrase our results. Theorem 1.2 . The knot polynomial invariants V_{1} , \Lambda_{1} and \Lambda_{-1} extend to invariants of links. The next theorem identifies the associated link invariants \Lambda_{1} and \Lambda_{-1} confirming a conjecture in [ GaKa ] . Theorem 1.3 . For all links L , we have \Lambda_{1,L}(t_{0},t_{1})=\Delta_{L}(t_{0})\,\Delta_{L}(t_{1})\in\mathbbm{Z}[t_{0}^{\pm 1},t_{1}^{\pm 1}], (1) and \Lambda_{-1,L}(t^{-2},s^{-2})=\Delta_{\mathfrak{sl}_{3},L}(t,s)\in\mathbbm{Z}[t^{\pm 2},s^{\pm 2}], (2) where \Delta is the (palindromic normalization of the) Alexander polynomial and \Delta_{\mathfrak{sl}_{3}} is the invariant studied in [ Harper ] . The Garoufalidis-Kashaev construction seems to offer a more natural setting for connecting quantum invariants to invariants from classical topology. Specifically, the coefficients of the R -matrices are valued in integral Laurent polynomials with canonical variables. In subsequent work, we will prove that the extension of V_{1} coincides with the Links–Gould invariant of links, verifying another conjecture of [ GaKa ] . We should mention that a local, tangle definition of these multivariable polynomials of links plays an important role in their efficient computation, as was explained time and again by Bar-Natan and van der Veen [ BNV:PG , BNV:API ] . In particular, it leads to an efficient computation of the V_{n} and the \Lambda_{\omega} polynomials of knots, that itself leads to newely discovered patterns in knot theory; see [ GL:patterns ] . 1.3. Plan of the proofs In Section 2 we recall the basics on tangles and the Reshetikhin–Turaev functor, as well as the properties of enhanced R -matrices that are used to realize the functor, and its comparison with the tangle construction of [ GaKa ] . Following Ohtsuki, we discuss in detail the Alexander R -matrix of a 2-dimensional space and how it leads to a link invariant, which is then identified with the Alexander polynomial. We then discuss the rigid R -matrix of V_{1} on a 4-dimensional space in Section 3 and produce an enhancement. We show that the enhanced R -matrix defines a link invariant. The extension of \Lambda_{1} to a link invariant and its identification with a product of Alexander polynomials is done simultaneously. This is proven by an appropriate conjugation of the enhanced R -matrix