Biquandle Fares and Link Invariants

Sam Nelson111Email: Sam.Nelson@cmc.edu. Partially supported by Simons Foundation collaboration grant 702597.    Stella Shah222Email: sshah0159@scrippscollege.edu
Abstract

We introduce a new family of invariants of oriented classical and virtual knots and links using fares, maps from paths in biquandle-colored diagrams to an abelian coefficient group. We consider the cases of 1-fares and 2-fares, provide examples to show that the enhancements are proper and end with some open questions about the cases of nn-fares for n>2n>2.

Keywords: biquandle fares, biquandles, quivers, link invariants

2020 MSC: 57K12

1 Introduction

Biquandles are algebraic structures providing solutions to the set-theoretic Yang-Baxter equation and invariants of oriented knots and links. More precisely, every oriented classical or virtual knot or link has a fundamental biquandle whose isomorphism class is a strong invariant and whose homsets into finite biquandles provide an infinite family of computable invariants. These homsets can be represented (and computed) via biquandle colorings of diagrams analogous to matrices representing homomorphisms in a module category.

The biquandle counting invariant is the cardinality of this homset. An enhancement of the biquandle counting invariant is a generally stronger invariant which determines the homset cardinality. In much previous work, enhancements have been defined by finding invariants of homset elements represented as coloring of diagrams and collecting the values of these invariants over the homset; infinite families of such examples include the biquandle 2-cocycle invariants [1, 2], biquandle bracket invariants [8, 5], biquandle coloring quiver enhancements [4, 6] and many more [9].

In this paper we introduce a new infinite family of enhancements of the biquandle homset invariant known as biquandle fares, inspired by the path algebra of a quiver (and the Tokyo subway system). A fare of order nn is a map from nn-tuples of elements of XX representing paths of length nn in a biquandle coloring of an oriented knot or link diagram interpreted as a directed graph to an abelian group AA. Summing the fare contributions over an appropriate set of paths yields a value in AA; the fare axioms are then the conditions required to ensure that this sum in not changed by XX-colored Reidemeister moves. Collecting the total fares over the homset yields the enhanced invariant in multiset form, which can be represented as a polynomial in various ways.

The paper is organized as follows. In Section 2 we review the basics of biquandles and homsets. In Section 3 we introduce the biquandle fare idea and illustrate with some toy examples. We handle the cases of fares of order 1,21,2 and 33 and discuss the case of higher order fares. For fares of order 2 and greater, we consider three basic types of fares: complete fares, through fares and crooked fares. We compute tables of invariant values for some examples choices of biquandle, coefficient ring and fare table. We conclude in Section 4 with some questions for future research.

This paper, including all text, illustrations, and python code for computations, was written strictly by the authors without the use of generative AI in any form.

2 Biquandles and Homsets

We begin by recalling a standard definition (see [3] for more).

Definition 1.

A biquandle is a set XX with two binary operations ¯,¯:X×XX\,\underline{\triangleright}\,,\,\overline{\triangleright}\,:X\times X\to X satisfying

  • (i)

    For all xXx\in X, x¯x=x¯xx\,\underline{\triangleright}\,x=x\,\overline{\triangleright}\,x,

  • (ii)

    For all yXy\in X, the maps αy,βy:XX\alpha_{y},\beta_{y}:X\to X defined by αy(x)=x¯y\alpha_{y}(x)=x\,\overline{\triangleright}\,y, βy(x)=x¯y\beta_{y}(x)=x\,\underline{\triangleright}\,y and S:X×XX×XS:X\times X\to X\times X defined by S(x,y)=(y¯x,x¯y)S(x,y)=(y\,\overline{\triangleright}\,x,x\,\underline{\triangleright}\,y) are invertible, and

  • (iii)

    For all x,y,zx,y,z, we have the exchange laws

    (x¯y)¯(z¯y)=(x¯z)¯(y¯z)(x¯y)¯(z¯y)=(x¯z)¯(y¯z)(x¯y)¯(z¯y)=(x¯z)¯(y¯z).\begin{array}[]{rcl}(x\,\underline{\triangleright}\,y)\,\underline{\triangleright}\,(z\,\underline{\triangleright}\,y)&=&(x\,\underline{\triangleright}\,z)\,\underline{\triangleright}\,(y\,\overline{\triangleright}\,z)\\ (x\,\underline{\triangleright}\,y)\,\overline{\triangleright}\,(z\,\underline{\triangleright}\,y)&=&(x\,\overline{\triangleright}\,z)\,\underline{\triangleright}\,(y\,\overline{\triangleright}\,z)\\ (x\,\overline{\triangleright}\,y)\,\overline{\triangleright}\,(z\,\overline{\triangleright}\,y)&=&(x\,\overline{\triangleright}\,z)\,\overline{\triangleright}\,(y\,\underline{\triangleright}\,z)\end{array}.

A biquandle in which x¯y=xx\,\overline{\triangleright}\,y=x for all x,yx,y is called a quandle.

Example 1.

Standard examples of biquandles include

  • Any group has biquandle structures given by

    x¯y=ynxynandx¯y=xx\,\underline{\triangleright}\,y=y^{-n}xy^{n}\quad\mathrm{and}\quad x\,\overline{\triangleright}\,y=x

    for any integer nn

  • Any group has biquandle structure given by

    x¯y=y1xy1andx¯y=x1x\,\underline{\triangleright}\,y=y^{-1}xy^{-1}\quad\mathrm{and}\quad x\,\overline{\triangleright}\,y=x^{-1}
  • Any module over a commutative unital ring RR has biquandle structures given by

    x¯y=tx+(st)yandx¯y=sxx\,\underline{\triangleright}\,y=tx+(s-t)y\quad\mathrm{and}\quad x\,\overline{\triangleright}\,y=sx

    for units t,sRt,s\in R.

The biquandle axioms are chosen so that for every XX-coloring of an oriented classical or virtual knot or link diagram (i.e., assignment of elements of XX to the semiarcs of DD satisfying the condition

[Uncaptioned image]

at every crossing) before a Reidemeister move, there is a unique XX-coloring of the diagram after the move which coincides with the original coloring outside the neighborhood of the move.

An XX-coloring determines and is determined by a unique biquandle homomorphism f:(L)Xf:\mathcal{B}(L)\to X from the fundamental biquandle of our oriented link LL to XX, subject to a choice of generating set; see [3, 7] for more. The set of XX-colorings of a diagram of LL is called the biquandle homset invariant Hom((L),X)\mathrm{Hom}(\mathcal{B}(L),X); its cardinality is known as the biquandle counting invariant, denoted ΦX(L)=|Hom((L),X)|\Phi_{X}^{\mathbb{Z}}(L)=|\mathrm{Hom}(\mathcal{B}(L),X)|.

Example 2.

Let XX be the biquandle given by the operation tables

¯123122221113333¯123123123123123\begin{array}[]{r|rrr}\,\underline{\triangleright}&1&2&3\\ \hline\cr 1&2&2&2\\ 2&1&1&1\\ 3&3&3&3\\ \end{array}\quad\begin{array}[]{r|rrr}\,\underline{\triangleright}&1&2&3\\ \hline\cr 1&2&3&1\\ 2&3&1&2\\ 3&1&2&3\end{array}

and consider the trefoil knot 313_{1}. There are three XX-colorings of 313_{1}, given by the homset

Hom(31,X)={[Uncaptioned image],[Uncaptioned image],[Uncaptioned image]}.\mathrm{Hom}(3_{1},X)=\left\{\raisebox{-39.74872pt}{\includegraphics{sn-ss2-9.pdf}},\ \raisebox{-39.74872pt}{\includegraphics{sn-ss2-10.pdf}},\ \raisebox{-39.74872pt}{\includegraphics{sn-ss2-11.pdf}}\right\}.

Hence we have ΦX(31)=3.\Phi_{X}^{\mathbb{Z}}(3_{1})=3.

3 Biquandle Fares

We begin with a definition.

Definition 2.

Let LL be an oriented classical or virtual knot or link diagram. A route of order nn in LL is an ordered sequence of nn semiarcs in LL such that the terminal crossing point of the kkth edge is the initial crossing point of the (k+1)(k+1)st edge for each k=1,,n1k=1,\dots,n-1. If we consider the underlying graph of the diagram LL as a quiver then routes are the paths in the quiver.

Let XX be a finite biquandle and AA and abelian group. A biquandle fare of order nn with coefficients in AA is a linear map ϕ:[Xn]A\phi:\mathbb{Z}[X^{n}]\to A satisfying axioms arising from the Reidemeister moves on XX-colored diagrams where we interpret an nn-tuple (x1,,xn)(x_{1},\dots,x_{n}) as corresponding to the biquandle colors on a route of order nn in a biquandle coloring of a diagram LL. More precisely, for each route of order nn we collect a fare contribution of (1)kϕ(x1,,xn)A(-1)^{k}\phi(x_{1},\dots,x_{n})\in A where kk is the number of negative crossings internal to the route.

[Uncaptioned image]+(1)2ϕ(x1,x2,x3,x4)\begin{array}[]{c}\includegraphics{sn-ss2-1.pdf}\\ +(-1)^{2}\phi(x_{1},x_{2},x_{3},x_{4})\end{array}

Given an XX-colored diagram, we find the set of all such contributions and take their sum in AA as the total fare of the diagram. The fare axioms are then chosen such that this sum of ϕ\phi-values in AA is invariant under XX-colored Reidemeister moves and hence defines an enhancement of the biquandle homset invariant.

The simplest case is fares of order n=1n=1.

Definition 3.

Let XX be a finite biquandle and AA an abelian group. Then a biquandle fare of order n=1n=1 with coefficients in AA is a function ϕ:XA\phi:X\to A satisfying the conditions

  • (i)

    For all xXx\in X, ϕ(x¯x)+ϕ(x)=0\phi(x\,\underline{\triangleright}\,x)+\phi(x)=0,

  • (ii)

    For all x,yXx,y\in X, ϕ(x)+ϕ(y)+ϕ(x¯y)+ϕ(y¯x)=0\phi(x)+\phi(y)+\phi(x\,\underline{\triangleright}\,y)+\phi(y\,\overline{\triangleright}\,x)=0 and

  • (iii)

    For all x,y,zXx,y,z\in X,

    ϕ(y)+ϕ(x¯y)+ϕ(x¯y)=ϕ(z¯x)+ϕ(x¯z)+ϕ((y¯x)¯(z¯x))\phi(y)+\phi(x\,\underline{\triangleright}\,y)+\phi(x\,\overline{\triangleright}\,y)=\phi(z\,\overline{\triangleright}\,x)+\phi(x\,\underline{\triangleright}\,z)+\phi((y\,\overline{\triangleright}\,x)\,\underline{\triangleright}\,(z\,\overline{\triangleright}\,x))

To compute the fare of an XX-colored diagram we can simply replace the biquandle colors xx on each edge with their values ϕ(x)\phi(x) in AA and sum them up. To see that the above conditions give invariance under Reidemeister moves it suffices to check each move in a generating set of oriented Reidemeister moves. In [10] several such sets are identified; we will use the (non-minimal) set consisting of all four moves of type I, all four moves of type II, and the all-positive move of type III.

Definition 4.

Let XX be a finite biquandle, DD an oriented knot or link diagram representing and oriented knot or link LL, AA an abelian group and ϕ\phi a biquandle fare. The multiset of of all appropriate fare values over the biquandle homset Hom((L),X)\mathrm{Hom}(\mathcal{B}(L),X) is the biquandle fare multiset, denoted ΦXϕ,M(L)\Phi_{X}^{\phi,M}(L).

Theorem 1.

Let XX be a biquandle, LL an oriented classical or virtual knot, link multi-knotoid, AA an abelian group and ϕ:XA\phi:X\to A a biquandle 11-fare. Then the multiset of fare values over the biquandle homset is invariant under Reidemeister moves.

Proof.

Taking Reidemeister I, we see that we need ϕ(x)+ϕ(x)+ϕ(x¯x)=ϕ(x)\phi(x)+\phi(x)+\phi(x\,\underline{\triangleright}\,x)=\phi(x) or more simply ϕ(x)+ϕ(x¯x)=0\phi(x)+\phi(x\,\underline{\triangleright}\,x)=0:

[Uncaptioned image]

The reader can check that the other three type I moves yield the same condition.

For Reidemeister II, we need

ϕ(x¯y)+ϕ(y)+ϕ(x¯y)+ϕ(y¯x)+ϕ(x)+ϕ(y)=ϕ(x¯y)+ϕ(y)\phi(x\,\underline{\triangleright}\,y)+\phi(y)+\phi(x\,\underline{\triangleright}\,y)+\phi(y\,\overline{\triangleright}\,x)+\phi(x)+\phi(y)=\phi(x\,\underline{\triangleright}\,y)+\phi(y)
[Uncaptioned image]

which is equivalent to ϕ(x)+ϕ(y)+ϕ(x¯y)+ϕ(y¯x)=0\phi(x)+\phi(y)+\phi(x\,\underline{\triangleright}\,y)+\phi(y\,\overline{\triangleright}\,x)=0. As in the type I case, the reader can verify that all four type II moves yield the same condition.

Finally, in the type III move we note that the colors on the outside of the move neighborhood yield the same sum on both sides of the move by the biquandle axioms, so we only need to consider the inner triangle contributions in the 11-fare case.

[Uncaptioned image].\includegraphics{sn-ss2-2.pdf}.

     

We can express a 1-fare for a biquandle X={x1,,xn}X=\{x_{1},\dots,x_{n}\} as an ordered nn-tuple ϕ=(ϕ(x1),,ϕ(xn))\phi=(\phi(x_{1}),\dots,\phi(x_{n})).

Example 3.

Let XX be the biquandle defined by the operation tables

¯123122121123333¯123122221113333.\begin{array}[]{r|rrr}\,\underline{\triangleright}&1&2&3\\ \hline\cr 1&2&2&1\\ 2&1&1&2\\ 3&3&3&3\end{array}\quad\begin{array}[]{r|rrr}\,\overline{\triangleright}&1&2&3\\ \hline\cr 1&2&2&2\\ 2&1&1&1\\ 3&3&3&3\end{array}.

Then we compute that XX has 11-fares over 2\mathbb{Z}_{2} given in the table

xϕ1(x)ϕ2(x)ϕ3(x)ϕ4(x)100112010130101.\begin{array}[]{r|cccc}x&\phi_{1}(x)&\phi_{2}(x)&\phi_{3}(x)&\phi_{4}(x)\\ \hline\cr 1&0&0&1&1\\ 2&0&1&0&1\\ 3&0&1&0&1\end{array}.

We will denote the 11-fare multiset invariant as ΦXϕ,M(L).\Phi_{X}^{\phi,M}(L). We can optionally express the multiset in one of several “polynomial” formats (scare quotes because these are not necessarily polynomials in the usual sense, depending on the choice of abelian group AA):

Definition 5.

Let XX be a biquandle, LL an oriented classical of virtual knot or link, AA and abelian group, ϕ:[Xn]A\phi:\mathbb{Z}[X^{n}]\to A a biquandle nn-fare and ΦXϕ,M(L)\Phi_{X}^{\phi,M}(L) the nn-fare multiset invariant. Then the additive nn-fare polynomial of LL with respect to (X,A,ϕ)(X,A,\phi) is

ΦXϕ,+(L)=fHom((L),X)xϕ(f)\Phi_{X}^{\phi,+}(L)=\sum_{f\in\mathrm{Hom}(\mathcal{B}(L),X)}x^{\phi(f)}

and the multiplicative nn-fare polynomial of LL with respect to (X,A,ϕ)(X,A,\phi) is

ΦXϕ,×(L)=fHom((L),X)(xϕ(f)).\Phi_{X}^{\phi,\times}(L)=\prod_{f\in\mathrm{Hom}(\mathcal{B}(L),X)}(x-\phi(f)).

The two polynomial forms are equivalent over most rings. For the additive form, evaluation at x=1x=1 yields the counting invariant value and the coefficient of a monomial tells us the number of colorings whose fare value is the monomial’s exponent, while in the multiplicative form the counting invariant is the degree of the polynomial and the fares are its roots.

Example 4.

Let XX be the biquandle defined by the operation tables

¯123131322223131¯123133322223111\begin{array}[]{r|rrr}\,\underline{\triangleright}&1&2&3\\ \hline\cr 1&3&1&3\\ 2&2&2&2\\ 3&1&3&1\end{array}\quad\begin{array}[]{r|rrr}\,\overline{\triangleright}&1&2&3\\ \hline\cr 1&3&3&3\\ 2&2&2&2\\ 3&1&1&1\\ \end{array}

and let A=22A=\mathbb{Z}_{2}\oplus\mathbb{Z}_{2} be the Klein 4-group. Then we compute that XX has 1-fares over AA including ϕ=((1,0),(0,1),(1,0))\phi=((1,0),(0,1),(1,0)). Then the virtual Hopf link LL has three XX-colorings

[Uncaptioned image]

Collecting the 11-fares over this homset yields the 11-fare multiset invariant value ΦXϕ,M(L)={2×(1,1),1×(0,0)}\Phi_{X}^{\phi,M}(L)=\{2\times(1,1),1\times(0,0)\}. We note that this distinguishes the virtual Hopf link from the unlink UU since a diagram without crossings has an empty set of routes and hence its multiset ΦXϕ,M(U)\Phi_{X}^{\phi,M}(U) can contain only zeros.

For n>1n>1, several cases are possible; we will consider three. For a complete fare we collect contributions from all possible routes of length nn. For a through fare we collect contributions from routes of length nn continuing along the knot either over or under at each crossing. For a crooked fare we collect contributions only from routes of length nn which turn at every vertex, never continuing across the crossing.

Definition 6.

Let XX be a finite biquandle and AA an abelian group. Then a complete biquandle fare of order n=2n=2 with coefficients in AA is a function ϕ:X2A\phi:X^{2}\to A satisfying the conditions

  • (i)

    For all xXx\in X,

    ϕ(x,x)+ϕ(x,x¯x)+ϕ(x¯x,x)+ϕ(x¯x,x¯x)=0,\phi(x,x)+\phi(x,x\,\underline{\triangleright}\,x)+\phi(x\,\underline{\triangleright}\,x,x)+\phi(x\,\underline{\triangleright}\,x,x\,\underline{\triangleright}\,x)=0,
  • (ii)

    For all x,yXx,y\in X,

    ϕ(x,y)+ϕ(x,x¯y)+ϕ(y¯x,y)+ϕ(y¯x,x¯y)ϕ(y,x)ϕ(x¯y,x)ϕ(y,y¯x)ϕ(x¯y,y¯x)=0\phi(x,y)+\phi(x,x\,\underline{\triangleright}\,y)+\phi(y\,\overline{\triangleright}\,x,y)+\phi(y\,\overline{\triangleright}\,x,x\,\underline{\triangleright}\,y)-\phi(y,x)-\phi(x\,\underline{\triangleright}\,y,x)-\phi(y,y\,\overline{\triangleright}\,x)-\phi(x\,\underline{\triangleright}\,y,y\,\overline{\triangleright}\,x)=0

    and

  • (iii)

    For all x,y,zXx,y,z\in X,

    ϕ(x,y)+ϕ(x,x¯y)+ϕ(y¯x,y)+ϕ(y¯x,x¯y)ϕ(x,z)+ϕ(x,x¯z)+ϕ(z¯x,z)+ϕ(z¯x,x¯z)+ϕ(y,z)+ϕ(y,y¯z)+ϕ(z¯y,z)+ϕ(z¯y,y¯z)+ϕ(y¯x,z¯x)+ϕ(y¯x,(y¯x)¯(z¯x))+ϕ(x¯y,z¯y)+ϕ(x¯y,(x¯y)¯(z¯y))=+ϕ((z¯x)¯(y¯x),z¯x)+ϕ((z¯x)¯(y¯x),(y¯x)¯(z¯x))+ϕ((z¯y)¯(x¯y),z¯y)+ϕ(x¯z,y¯z)+ϕ(x¯z,(x¯z)¯(y¯z))+ϕ((z¯y)¯(x¯y),(x¯y)¯(z¯y))+ϕ((y¯z)¯(x¯z),y¯z)+ϕ((y¯z)¯(x¯z),(x¯z)¯(y¯z))\begin{array}[]{rcl}\phi(x,y)+\phi(x,x\,\underline{\triangleright}\,y)+\phi(y\,\overline{\triangleright}\,x,y)+\phi(y\,\overline{\triangleright}\,x,x\,\underline{\triangleright}\,y)&&\phi(x,z)+\phi(x,x\,\underline{\triangleright}\,z)+\phi(z\,\overline{\triangleright}\,x,z)+\phi(z\,\overline{\triangleright}\,x,x\,\underline{\triangleright}\,z)\\ +\phi(y,z)+\phi(y,y\,\underline{\triangleright}\,z)+\phi(z\,\overline{\triangleright}\,y,z)+\phi(z\,\overline{\triangleright}\,y,y\,\underline{\triangleright}\,z)&&+\phi(y\,\overline{\triangleright}\,x,z\,\overline{\triangleright}\,x)+\phi(y\,\overline{\triangleright}\,x,(y\,\overline{\triangleright}\,x)\,\underline{\triangleright}\,(z\,\overline{\triangleright}\,x))\\ +\phi(x\,\underline{\triangleright}\,y,z\,\overline{\triangleright}\,y)+\phi(x\,\underline{\triangleright}\,y,(x\,\underline{\triangleright}\,y)\,\underline{\triangleright}\,(z\,\overline{\triangleright}\,y))&=&+\phi((z\,\overline{\triangleright}\,x)\,\overline{\triangleright}\,(y\,\overline{\triangleright}\,x),z\,\overline{\triangleright}\,x)+\phi((z\,\overline{\triangleright}\,x)\,\overline{\triangleright}\,(y\,\overline{\triangleright}\,x),(y\,\overline{\triangleright}\,x)\,\underline{\triangleright}\,(z\,\overline{\triangleright}\,x))\\ +\phi((z\,\overline{\triangleright}\,y)\,\overline{\triangleright}\,(x\,\underline{\triangleright}\,y),z\,\overline{\triangleright}\,y)&&+\phi(x\,\underline{\triangleright}\,z,y\,\underline{\triangleright}\,z)+\phi(x\,\underline{\triangleright}\,z,(x\,\underline{\triangleright}\,z)\,\underline{\triangleright}\,(y\,\underline{\triangleright}\,z))\\ +\phi((z\,\overline{\triangleright}\,y)\,\overline{\triangleright}\,(x\,\underline{\triangleright}\,y),(x\,\underline{\triangleright}\,y)\,\underline{\triangleright}\,(z\,\overline{\triangleright}\,y))&&+\phi((y\,\underline{\triangleright}\,z)\,\overline{\triangleright}\,(x\,\underline{\triangleright}\,z),y\,\underline{\triangleright}\,z)+\phi((y\,\underline{\triangleright}\,z)\,\overline{\triangleright}\,(x\,\underline{\triangleright}\,z),(x\,\underline{\triangleright}\,z)\,\underline{\triangleright}\,(y\,\underline{\triangleright}\,z))\\ \end{array}

Example 5.

Let XX be the biquandle given by the operation tables

¯12122211¯12122211.\begin{array}[]{r|rr}\,\underline{\triangleright}&1&2\\ \hline\cr 1&2&2\\ 2&1&1\end{array}\quad\begin{array}[]{r|rr}\,\overline{\triangleright}&1&2\\ \hline\cr 1&2&2\\ 2&1&1\end{array}.

Then XX has complete 2-fares over A=5A=\mathbb{Z}_{5} including

ϕ12104242.\begin{array}[]{r|rr}\phi&1&2\\ \hline\cr 1&0&4\\ 2&4&2\end{array}.

The Hopf link L2a1L2a1 has four XX-colorings

[Uncaptioned image]

and each coloring has a total of eight routes of order 2, resulting in a multiset of four fares:

ΦXϕ,M={ϕ(1,1)+ϕ(1,2)+ϕ(2,1)+ϕ(2,2)+ϕ(1,1)+ϕ(1,2)+ϕ(2,1)+ϕ(2,2),ϕ(1,2)+ϕ(1,2)+ϕ(1,2)+ϕ(1,2)+ϕ(2,1)+ϕ(2,1)+ϕ(2,1)+ϕ(2,1),ϕ(2,1)+ϕ(2,1)+ϕ(2,1)+ϕ(2,1)+ϕ(1,2)+ϕ(1,2)+ϕ(1,2)+ϕ(1,2),ϕ(1,1)+ϕ(1,2)+ϕ(2,1)+ϕ(2,2)+ϕ(1,1)+ϕ(1,2)+ϕ(2,1)+ϕ(2,2),}={0+4+4+2+0+4+4+2, 4+4+4+4+4+4+4+4,4+4+4+4+4+4+4+4, 0+4+4+2+0+4+4+2}={0,2,2,0}\begin{array}[]{rcl}\Phi_{X}^{\phi,M}&=&\{\phi(1,1)+\phi(1,2)+\phi(2,1)+\phi(2,2)+\phi(1,1)+\phi(1,2)+\phi(2,1)+\phi(2,2),\\ &&\phi(1,2)+\phi(1,2)+\phi(1,2)+\phi(1,2)+\phi(2,1)+\phi(2,1)+\phi(2,1)+\phi(2,1),\\ &&\phi(2,1)+\phi(2,1)+\phi(2,1)+\phi(2,1)+\phi(1,2)+\phi(1,2)+\phi(1,2)+\phi(1,2),\\ &&\phi(1,1)+\phi(1,2)+\phi(2,1)+\phi(2,2)+\phi(1,1)+\phi(1,2)+\phi(2,1)+\phi(2,2),\}\\ &=&\{0+4+4+2+0+4+4+2,\ 4+4+4+4+4+4+4+4,\\ &&4+4+4+4+4+4+4+4,\ 0+4+4+2+0+4+4+2\}\\ &=&\{0,2,2,0\}\end{array}

with polynomial versions ΦXϕ,+(L2a1)=2+x2\Phi_{X}^{\phi,+}(L2a1)=2+x^{2} and ΦXϕ,×(L2a1)=x2(x2)2\Phi_{X}^{\phi,\times}(L2a1)=x^{2}(x-2)^{2}.

Example 6.

Let XX be the biquandle given by the operation tables

¯123122121123333¯123122221113333.\begin{array}[]{r|rrr}\,\underline{\triangleright}&1&2&3\\ \hline\cr 1&2&2&1\\ 2&1&1&2\\ 3&3&3&3\end{array}\quad\begin{array}[]{r|rrr}\,\overline{\triangleright}&1&2&3\\ \hline\cr 1&2&2&2\\ 2&1&1&1\\ 3&3&3&3\end{array}.

Then we compute via python that XX has complete 2-fares over 5\mathbb{Z}_{5} including

ϕ=[003003110]\phi=\left[\begin{array}[]{rrr}0&0&3\\ 0&0&3\\ 1&1&0\end{array}\right]

(where the entry in row jj column kk is ϕ(j,k)\phi(j,k)). Then the multiplicative polynomial fare values for the classical links with up to seven crossings are given in the table.

LΦXϕ,×(L)L2a1x5L4a1x5(x1)4L5a1x9L6a1x5(x1)4L6a2x5L6a3x5L6a4x9(x1)18L6a5x9(x1)6L6n1x9(x1)6LΦXϕ,×(L)L7a1x9L7a2x5(x1)4L7a3x9L7a4x9L7a5x5L7a6x5L7a7x13(x1)2L7n1x5(x4)4L7n2x9\begin{array}[]{r|l}L&\Phi_{X}^{\phi,\times}(L)\\ \hline\cr L2a1&x^{5}\\ L4a1&x^{5}(x-1)^{4}\\ L5a1&x^{9}\\ L6a1&x^{5}(x-1)^{4}\\ L6a2&x^{5}\\ L6a3&x^{5}\\ L6a4&x^{9}(x-1)^{18}\\ L6a5&x^{9}(x-1)^{6}\\ L6n1&x^{9}(x-1)^{6}\\ \end{array}\quad\begin{array}[]{r|l}L&\Phi_{X}^{\phi,\times}(L)\\ \hline\cr L7a1&x^{9}\\ L7a2&x^{5}(x-1)^{4}\\ L7a3&x^{9}\\ L7a4&x^{9}\\ L7a5&x^{5}\\ L7a6&x^{5}\\ L7a7&x^{13}(x-1)^{2}\\ L7n1&x^{5}(x-4)^{4}\\ L7n2&x^{9}\\ \end{array}

Next we have another type of 2-fare.

Definition 7.

Let XX be a finite biquandle and AA an abelian group. Then a through biquandle fare of order n=2n=2 with coefficients in AA is a function ϕ:X2A\phi:X^{2}\to A satisfying the conditions

  • (i)

    For all xXx\in X,

    ϕ(x,x¯x)+ϕ(x¯x,x)=0,\phi(x,x\,\underline{\triangleright}\,x)+\phi(x\,\underline{\triangleright}\,x,x)=0,
  • (ii)

    For all x,yXx,y\in X,

    ϕ(x,x¯y)+ϕ(y¯x,y)ϕ(x¯y,x)ϕ(y,y¯x)=0\phi(x,x\,\underline{\triangleright}\,y)+\phi(y\,\overline{\triangleright}\,x,y)-\phi(x\,\underline{\triangleright}\,y,x)-\phi(y,y\,\overline{\triangleright}\,x)=0

    and

  • (iii)

    For all x,y,zXx,y,z\in X,

    ϕ(x,x¯y)+ϕ(y¯x,y)ϕ(x,x¯z)+ϕ(z¯x,z)+ϕ(y,y¯z)+ϕ(z¯y,z)=+ϕ(y¯x,(y¯x)¯(z¯x))+ϕ((z¯x)¯(y¯x),z¯x)+ϕ(x¯y,(x¯y)¯(z¯y))+ϕ((z¯y)¯(x¯y),z¯y)+ϕ(x¯z,(x¯z)¯(y¯z))+ϕ((y¯z)¯(x¯z),y¯z)\begin{array}[]{rcl}\phi(x,x\,\underline{\triangleright}\,y)+\phi(y\,\overline{\triangleright}\,x,y)&&\phi(x,x\,\underline{\triangleright}\,z)+\phi(z\,\overline{\triangleright}\,x,z)\\ +\phi(y,y\,\underline{\triangleright}\,z)+\phi(z\,\overline{\triangleright}\,y,z)&=&+\phi(y\,\overline{\triangleright}\,x,(y\,\overline{\triangleright}\,x)\,\underline{\triangleright}\,(z\,\overline{\triangleright}\,x))+\phi((z\,\overline{\triangleright}\,x)\,\overline{\triangleright}\,(y\,\overline{\triangleright}\,x),z\,\overline{\triangleright}\,x)\\ +\phi(x\,\underline{\triangleright}\,y,(x\,\underline{\triangleright}\,y)\,\underline{\triangleright}\,(z\,\overline{\triangleright}\,y))+\phi((z\,\overline{\triangleright}\,y)\,\overline{\triangleright}\,(x\,\underline{\triangleright}\,y),z\,\overline{\triangleright}\,y)&&+\phi(x\,\underline{\triangleright}\,z,(x\,\underline{\triangleright}\,z)\,\underline{\triangleright}\,(y\,\underline{\triangleright}\,z))+\phi((y\,\underline{\triangleright}\,z)\,\overline{\triangleright}\,(x\,\underline{\triangleright}\,z),y\,\underline{\triangleright}\,z)\\ \end{array}

Example 7.

Let XX be the biquandle (indeed, quandle) defined by the operation tables

¯123411342242133243143124¯123411111222223333344444\begin{array}[]{r|rrrr}\,\underline{\triangleright}&1&2&3&4\\ \hline\cr 1&1&3&4&2\\ 2&4&2&1&3\\ 3&2&4&3&1\\ 4&3&1&2&4\end{array}\quad\begin{array}[]{r|rrrr}\,\overline{\triangleright}&1&2&3&4\\ \hline\cr 1&1&1&1&1\\ 2&2&2&2&2\\ 3&3&3&3&3\\ 4&4&4&4&4\end{array}

Our python computations reveal 125 through 2-fares with 5\mathbb{Z}_{5} coefficients including

ϕ123410030200303220240030\begin{array}[]{r|rrrr}\phi&1&2&3&4\\ \hline\cr 1&0&0&3&0\\ 2&0&0&3&0\\ 3&2&2&0&2\\ 4&0&0&3&0\end{array}

where the entry in row jj column kk is ϕ(j,k)\phi(j,k).

The figure eight knot 414_{1} has sixteen XX-colorings including

[Uncaptioned image].\includegraphics{sn-ss2-5.pdf}.

Each of the eight semi-arc starts a length-2 through path; summing the contributions, we obtain a total biquandle through fare of

ϕ(2,2)ϕ(2,1)ϕ(1,1)+ϕ(1,3)+ϕ(3,3)ϕ(3,4)ϕ(4,4)+ϕ(4,2)=000+3+020+0=1.\phi(2,2)-\phi(2,1)-\phi(1,1)+\phi(1,3)+\phi(3,3)-\phi(3,4)-\phi(4,4)+\phi(4,2)=0-0-0+3+0-2-0+0=1.

Repeating for the other 15 colorings, we obtain biquandle 2-fare multiset and polynomial invariant values

ΦXϕ,M(41)={4×0,6×1,6×4},ΦXϕ,+(41)=4+6x+6x4andΦXϕ,×=x4(x1)6(x4)4.\Phi_{X}^{\phi,M}(4_{1})=\{4\times 0,6\times 1,6\times 4\},\ \Phi_{X}^{\phi,+}(4_{1})=4+6x+6x^{4}\ \mathrm{and}\ \Phi_{X}^{\phi,\times}=x^{4}(x-1)^{6}(x-4)^{4}.

This example shows that these invariants are proper enhancements of the (bi)quandle counting invariant since the trefoil knot 313_{1} has the same counting invariant value of 1616 with respect to XX but has trivial values of the enhanced invariants,

ΦXϕ,M={16×0},ΦXϕ,+(01)=16andΦXϕ,×(01)=x16.\Phi_{X}^{\phi,M}=\{16\times 0\},\ \Phi_{X}^{\phi,+}(0_{1})=16\ \mathrm{and}\ \Phi_{X}^{\phi,\times}(0_{1})=x^{16}.
Example 8.

Let XX be the biquandle given by the operation tables

¯123411342224313312444213¯123411111244443222343333.\begin{array}[]{r|rrrr}\,\underline{\triangleright}&1&2&3&4\\ \hline\cr 1&1&3&4&2\\ 2&2&4&3&1\\ 3&3&1&2&4\\ 4&4&2&1&3\end{array}\quad\begin{array}[]{r|rrrr}\,\underline{\triangleright}&1&2&3&4\\ \hline\cr 1&1&1&1&1\\ 2&4&4&4&4\\ 3&2&2&2&3\\ 4&3&3&3&3\end{array}.

Then we compute that XX has through fares over 6\mathbb{Z}_{6} including

ϕ123413141250333230045303\begin{array}[]{r|rrrr}\phi&1&2&3&4\\ \hline\cr 1&3&1&4&1\\ 2&5&0&3&3\\ 3&2&3&0&0\\ 4&5&3&0&3\end{array}

and the multiplicative polynomial values for the prime classical knots with up to eight crossings in the table.

ΦXϕ,×(K)Kx451,52,61,62,63,71,74,75,7677,82,83,86,87,88,89,812,814,816,817x4(x2)6(x4)641,81,811x1631,72,73,84,85,810,813,815,819,820,821x4(x2)30(x4)30818.\begin{array}[]{r|l}\Phi_{X}^{\phi,\times}(K)&K\\ \hline\cr x^{4}&5_{1},5_{2},6_{1},6_{2},6_{3},7_{1},7_{4},7_{5},7_{6}7_{7},8_{2},8_{3},8_{6},8_{7},8_{8},8_{9},8_{12},8_{14},8_{16},8_{17}\\ x^{4}(x-2)^{6}(x-4)^{6}&4_{1},8_{1},8_{11}\\ x^{16}&3_{1},7_{2},7_{3},8_{4},8_{5},8_{10},8_{13},8_{15},8_{19},8_{20},8_{21}\\ x^{4}(x-2)^{30}(x-4)^{30}&8_{18}\end{array}.

In particular, we note that the invariant distinguishes several prime classical knots with equal counting invariant values and hence is a proper enhancement.

Finally, we define a third type of 2-fare.

Definition 8.

Let XX be a finite biquandle and AA an abelian group. Then a crooked biquandle fare of order n=2n=2 with coefficients in AA is a function ϕ:X2A\phi:X^{2}\to A satisfying the conditions

  • (i)

    For all xXx\in X, ϕ(x,x¯x)+ϕ(x¯x,x)=0\phi(x,x\,\underline{\triangleright}\,x)+\phi(x\,\underline{\triangleright}\,x,x)=0,

  • (ii)

    For all x,yXx,y\in X,

    ϕ(x,y)+ϕ(y¯x,x¯y)ϕ(y,x)ϕ(x¯y,y¯x)=0\phi(x,y)+\phi(y\,\overline{\triangleright}\,x,x\,\underline{\triangleright}\,y)-\phi(y,x)-\phi(x\,\underline{\triangleright}\,y,y\,\overline{\triangleright}\,x)=0

    and

  • (iii)

    For all x,y,zXx,y,z\in X,

    ϕ(x,y)+ϕ(y¯x,x¯y)ϕ(x,z)+ϕ(z¯x,x¯z)+ϕ(y,z)+ϕ(z¯y,y¯z)=+ϕ(y¯x,z¯x)+ϕ((z¯x)¯(y¯x),(y¯x)¯(z¯x))+ϕ(x¯y,z¯y)+ϕ((z¯y)¯(x¯y),(x¯y)¯(z¯y))+ϕ(x¯z,y¯z)+ϕ((y¯z)¯(x¯z),(x¯z)¯(y¯z))\begin{array}[]{rcl}\phi(x,y)+\phi(y\,\overline{\triangleright}\,x,x\,\underline{\triangleright}\,y)&&\phi(x,z)+\phi(z\,\overline{\triangleright}\,x,x\,\underline{\triangleright}\,z)\\ +\phi(y,z)+\phi(z\,\overline{\triangleright}\,y,y\,\underline{\triangleright}\,z)&=&+\phi(y\,\overline{\triangleright}\,x,z\,\overline{\triangleright}\,x)+\phi((z\,\overline{\triangleright}\,x)\,\overline{\triangleright}\,(y\,\overline{\triangleright}\,x),(y\,\overline{\triangleright}\,x)\,\underline{\triangleright}\,(z\,\overline{\triangleright}\,x))\\ +\phi(x\,\underline{\triangleright}\,y,z\,\overline{\triangleright}\,y)+\phi((z\,\overline{\triangleright}\,y)\,\overline{\triangleright}\,(x\,\underline{\triangleright}\,y),(x\,\underline{\triangleright}\,y)\,\underline{\triangleright}\,(z\,\overline{\triangleright}\,y))&&+\phi(x\,\underline{\triangleright}\,z,y\,\underline{\triangleright}\,z)+\phi((y\,\underline{\triangleright}\,z)\,\overline{\triangleright}\,(x\,\underline{\triangleright}\,z),(x\,\underline{\triangleright}\,z)\,\underline{\triangleright}\,(y\,\underline{\triangleright}\,z))\\ \end{array}

As with fares of type n=1n=1, we have:

Theorem 2.

Let XX be a biquandle, LL an oriented classical or virtual link, AA an abelian group and ϕ:XA\phi:X\to A a complete, through or crooked biquandle 22-fare. Then the multiset of fare values over the biquandle homset is invariant under Reidemeister moves.

Proof.

The proof is a check of invariance under Reidemeister moves analogous to the proof of Theorem 1.      

For the case of fares of order 22, a choice of through fare ϕ1\phi_{1} and crooked fare ϕ2\phi_{2} add to yield a complete fare ϕ1+ϕ2\phi_{1}+\phi_{2}. The converse is not true in general – a complete fare need not decompose as a sum of a through fare and a crooked fare. We will say a complete 2-fare is decomposable if it is a sum of a crooked fare and a through fare and indecomposable otherwise.

[Uncaptioned image]
Example 9.

Let XX be the biquandle given by the operation tables

¯123411342224133132444231¯123411342224133132444231\begin{array}[]{r|rrrr}\,\underline{\triangleright}&1&2&3&4\\ \hline\cr 1&1&3&4&2\\ 2&2&4&1&3\\ 3&1&3&2&4\\ 4&4&2&3&1\end{array}\quad\begin{array}[]{r|rrrr}\,\overline{\triangleright}&1&2&3&4\\ \hline\cr 1&1&3&4&2\\ 2&2&4&1&3\\ 3&1&3&2&4\\ 4&4&2&3&1\end{array}

We use Python to compute that there are 3125 crooked biquandle 2-fares over 5\mathbb{Z}_{5}. One such example is

ϕ123411342231103024241434\begin{array}[]{r|rrrr}\phi&1&2&3&4\\ \hline\cr 1&1&3&4&2\\ 2&3&1&1&0\\ 3&0&2&4&2\\ 4&1&4&3&4\end{array}

where, as in Example 7, ϕ(j,k)\phi(j,k) is the entry in the kkth column of the jjth row. There are 64 total X-colorings of the link L7a7, including the following:

[Uncaptioned image].\includegraphics{sn-ss2-13.pdf}.

There are 7 crossings, and thus, 14 distinct crooked 2-routes in each XX-coloring. Summing the fares over these, we obtain the crooked fare of this coloring to be:

2routesϕ(x,y)\displaystyle\sum_{\mathrm{2-routes}}\phi(x,y) =\displaystyle= ϕ(2,4)+ϕ(3,2)+ϕ(3,1)+ϕ(3,2)+ϕ(2,3)+ϕ(1,3)+ϕ(4,2)\displaystyle\phi(2,4)+\phi(3,2)+\phi(3,1)+\phi(3,2)+\phi(2,3)+\phi(1,3)+\phi(4,2)
+ϕ(2,3)ϕ(3,1)ϕ(3,2)ϕ(1,3)ϕ(2,3)ϕ(4,4)ϕ(3,3)\displaystyle+\phi(2,3)-\phi(3,1)-\phi(3,2)-\phi(1,3)-\phi(2,3)-\phi(4,4)-\phi(3,3)
=\displaystyle= 4+1+4+1+2+0+0+2410244\displaystyle 4+1+4+1+2+0+0+2-4-1-0-2-4-4
=\displaystyle= 4.\displaystyle 4.

Repeating for the other 63 colorings, we find that 12 have a fare of 4, 24 have a fare of 2, and the remaining 36 have a fare of 0. Thus, we have the multiplicative polynomial invariant value

ΦXϕ,×(L7a7)=x36(x2)24(x4)12.\Phi_{X}^{\phi,\times}(L7a7)=x^{36}(x-2)^{24}(x-4)^{12}.

4 Questions

We conclude with some questions and directions for future research.

It is important first of all to note that the examples we have computed here were chosen as small easily computable toy examples and should not be taken as representative of the power of this infinite family of invariants. Our examples use small cardinality biquandles with small cardinality finite ring coefficients; larger biquandles and larger finite and infinite coefficient rings are expected to yield stronger invariants.

We have considered only the cases of 1-fares and 2-fares; the next cases to tackle should be 3-fares, 4-fares, et cetera. These cases pose a technical challenge in that to find all possible cases we must expand the neighborhood of each move to include more crossings. Nonetheless, nn-fares are definable for every natural number nn.

Finally, we note that a crooked 2-fare is very similar to a 2-cocycle; this inspires the question of what is the algebraic structure of the set of all nn-fares of a particular type (crooked, complete, through, etc.) for a given choice of biquandle and coefficient ring. Is there an underlying homology theory as with quandle/Yang-Baxter homology? What is the relation, if any, with biquandle brackets?

References

  • [1] J. S. Carter, D. Jelsovsky, S. Kamada, L. Langford, and M. Saito. State-sum invariants of knotted curves and surfaces from quandle cohomology. Electron. Res. Announc. Amer. Math. Soc., 5:146–156 (electronic), 1999.
  • [2] J. Ceniceros, M. Elhamdadi, M. Green, and S. Nelson. Augmented biracks and their homology. Internat. J. Math., 25(9):1450087, 19, 2014.
  • [3] M. Elhamdadi and S. Nelson. Quandles—an introduction to the algebra of knots, volume 74 of Student Mathematical Library. American Mathematical Society, Providence, RI, 2015.
  • [4] P. C. Falkenburg and S. Nelson. Biquandle bracket quivers. J. Knot Theory Ramifications (to appear), 2023.
  • [5] N. Gügümcü, S. Nelson, and N. Oyamaguchi. Biquandle brackets and knotoids. J. Knot Theory Ramifications, 30(9):Paper No. 2150064, 17, 2021.
  • [6] Y. Joung and S. Nelson. Bikei module invariants of unoriented surface-links. J. Knot Theory Ramifications, 32(10):Paper No. 2350065, 14, 2023.
  • [7] S. Nelson. A quick note on homsets and diagrams. J. Knot Theory Ramifications (to appear), 2025.
  • [8] S. Nelson, M. E. Orrison, and V. Rivera. Quantum enhancements and biquandle brackets. J. Knot Theory Ramifications, 26(5):1750034, 24, 2017.
  • [9] S. Nelson and E. Watterberg. Birack dynamical cocycles and homomorphism invariants. J. Algebra Appl., 12(8):1350049, 14, 2013.
  • [10] M. Polyak. Minimal generating sets of Reidemeister moves. Quantum Topol., 1(4):399–411, 2010.

Department of Mathematical Sciences
Claremont McKenna College
850 Columbia Ave.
Claremont, CA 91711