Shadow biquandles and local biquandles
Kanako Oshiro

TL;DR
This paper explores the relationship between shadow biquandles and local biquandles, showing their (co)homology groups and cocycle invariants are equivalent, and connects these structures to knot theory and existing cocycle theories.
Contribution
It establishes an isomorphism between shadow biquandle and local biquandle (co)homology groups and relates cocycle invariants to Niebrzydowski's theory and Mochizuki's cocycles.
Findings
(Co)homology groups of shadow biquandles are isomorphic to those of local biquandles.
Cocycle invariants for links and surface-links coincide between shadow and local biquandles.
Certain cocycles can be induced from Mochizuki's cocycles.
Abstract
Given a shadow biquandle composed of a biquandle and a strongly connected -set , we have a local biquandle structure on . The (co)homology groups of such shadow biquandles are isomorphic to those of the corresponding local biquandles. Moreover, cocycle invariants, of oriented links and oriented surface-links, using such shadow biquandles coincide with those using the corresponding local biquandles. These results imply that for some cases, the Niebrzydowski's theory in [14, 15, 16] for knot-theoretic ternary quasigroups is the same as shadow biquandle theory. We also show that some local biquandle - or -cocycles and some - or -cocycles of the Niebrzydowski's (co)homology theory can be induced from Mochizuki's cocycles.
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
Taxonomy
TopicsGeometric and Algebraic Topology · Advanced Combinatorial Mathematics · Homotopy and Cohomology in Algebraic Topology
Shadow biquandles and local biquandles
Kanako Oshiro
Department of Information and Communication Sciences, Sophia University, Tokyo 102-8554, Japan
Abstract.
Given a shadow biquandle composed of a biquandle and a strongly connected -set , we have a local biquandle structure on . The (co)homology groups of such shadow biquandles are isomorphic to those of the corresponding local biquandles. Moreover, cocycle invariants, of oriented links and oriented surface-links, using such shadow biquandles coincide with those using the corresponding local biquandles. These results imply that for some cases, the Niebrzydowski’s theory in [14, 15, 16] for knot-theoretic ternary quasigroups is the same as shadow biquandle theory. We also show that some local biquandle - or -cocycles and some - or -cocycles of the Niebrzydowski’s (co)homology theory can be induced from Mochizuki’s cocycles.
Key words and phrases:
Shadow biquandles, local biquandles, homology groups, cohomology groups, cocycle invariants, links and surface-links
2010 Mathematics Subject Classification:
57M27, 57M25
1. Introduction
In knot theory and related topics, quandles [7, 11] are important algebraic systems, each of which equips a binary operation coming from Reidemeister moves of oriented link diagrams with arc labelings. Biquandles [4, 9] are a generalization of quandles and they are also important algebraic systems, each of which equips two binary operations coming from Reidemeister moves of oriented link diagrams with semi-arc labelings. As other (or further) important generalizations, shadow quandle theory and shadow biquandle theory were introduced and well-studied, see [2, 5, 8] for example. Both of them are related to region labelings for oriented link diagrams in addition to arc or semi-arc labelings. In each of quandle theory, biquandle theory, shadow quandle theory and shadow biquandle theory, a (co)homology theory and a cocycle invariant of oriented links (or oriented surface-links, oriented virtual-links and so on) using a cocycle of the (co)homology theory are defined and well-studied.
In [14, 15, 16], Niebrzydowski studied an algebraic system, called a knot-theoretic ternary quasigroup, which equips a ternary operation coming from Reidemeister moves of oriented link diagrams with region labelings. He defined a (co)homology theory of the algebraic systems and a cocycle invariant of oriented links and oriented surface-links using a cocycle of his (co)homology theory, see also [3, 10, 13]. Note that the region labelings in this case are not related to arc or semi-arc labelings, while the region labelings in the cases of shadow quandles or shadow biquandles depend on the arc or semi-arc labelings.
In [17], local biquandle theory was introduced. A local biquandle is always given associated with a knot-theoretic (horizontal- or vertical-)ternary-quasigroup. Although a local biquandle is not a biquandle, it has a local algebraic structure that is similar to the algebraic structure of biquandles, that is, it has a local algebraic structure related to semi-arc labelings. It was shown that the Niebrzydowski’s (co)homology theory can be interpreted as local biquandle theory. On other words, in some sense, the Niebrzydowski’s (co)homology theory can be interpreted similarly as biquandle (co)homology theory since local biquandle (co)homology theory is an analogy of biquandle (co)homology theory. Furthermore, through the interpretation of the (co)homology theories, it was shown that the Niebrzydowski’s cocycle invariants and the local biquandle cocycle invariants of oriented links and oriented surface-links are the same. This implies that in some sense, the Niebrzydowski’s cocycle invariants can be also interpreted similarly as the biquandle cocycle invariants.
In this paper, we show that given a shadow biquandle with a strongly connected -set , we can define a knot-theoretic (horizontal-)ternary-quasigroup (see Theorem 4.1), and then, we have a local biquandle associated with . The (co)homology groups of are isomorphic to those of the corresponding local biquandle (see Theorem 4.4), and the shadow biquandle cocycle invariant using and a cocycle coincides with the local biquandle cocycle invariant using the corresponding local biquandle and a corresponding cocycle (see Theorems 4.6 and 4.8). Considering the main results in this paper together with the results shown in [17], we can say that for some cases, the Niebrzydowski’s theory in [14, 15, 16] is the same as shadow biquandle theory (see Corollaries 5.1, 5.2 and 5.3). As a consequence, we show that some local biquandle - or -cocycles and some - or -cocycles of the Niebrzydowski’s (co)homology theory can be induced from Mochizuki’s cocycles in [12] (see Examples 5.4).
The paper is organized as follows: In Section 2, we review the definitions of links, surface-links, biquandles, horizontal-tribrackets and local biquandles. In Section 3, we recall the definitions of shadow biquandle (co)homology groups, shadow biquandle cocycle invariants, local biquandle (co)homology groups and local biquandle cocycle invariants. The main results (Theorems 4.1, 4.4, 4.6 and 4.8) in this paper are stated and proven in Section 4. In Section 5, we give a relationship between the results in this paper and the Niebrzydowski’s theory given in [14, 15, 16], and we show some local biquandle cocycles and some cocycles of the Niebrzydowski’s (co)homology theory that are induced from Mochizuki’s cocycles.
2. Preliminaries
2.1. Links, surface-links, connected diagrams
A knot is an oriented -dimensional sphere embedded in . A link is a disjoint union of knots. We note that every knot is a link. Two links are said to be equivalent if they can be deformed into each other through an isotopy of . A diagram of a link is its image by a regular projection, from to , equipped with the height information for each double point. It is known that two link diagrams represent the same link if and only if they are related by a finite sequence of Reidemeister moves. A knot diagram is always connected. A link diagram with at least two components is said to be connected if every component intersects another component. It is known that between two connected diagrams and that represent the same link, there exists a finite sequence of connected diagrams and oriented Reidemeister moves that transforms to , i.e., there exists
[TABLE]
where for each , is an oriented Reidemeister move, and is a connected diagram of a link. For a diagram , we remove a small neighborhood of each crossing, and then, we call each connected component a semi-arc of . In this paper, for a link diagram , means the set of semi-arcs of and means the set of connected regions of . For a semi-arc of a link diagram , we assign a normal vector to to satisfy that the pair of the orientation of and coincides with the right-handed orientation of , and thus, we represent the orientation of .
A surface-knot is an oriented closed surface locally flatly embedded in . A surface-link is a disjoint union of surface-knots. We note that every surface-knot is a surface-link. Two surface-links are said to be equivalent if they can be deformed into each other through an isotopy of . A diagram of a surface-link is its image by a regular projection, from to , equipped with the height information for each double point curve, where the height information is represented by removing small neighborhoods of lower double point curves. Then a diagram is composed of four kinds of local pictures depicted in Figure 1, and the indicated points are called a regular point, a double point, a triple point and a branch point, respectively. It is known that two surface-link diagrams represent the same surface-link if and only if they are related by a finite sequence of Roseman moves, see [18] for details. A surface-knot diagram is always connected. A surface-link diagram with at least two components is said to be connected if every component intersects another component. It is known that between two connected diagrams and that represent the same surface-link, there exists a finite sequence of connected diagrams and oriented Roseman moves that transforms to . For a surface-link diagram , we remove small neighborhoods of double point curves, and then, we call each connected component a semi-sheet of . In this paper, for a surface-link diagram , means the set of semi-sheets of and means the set of connected regions of . For a semi-sheet of a surface-link diagram , we assign a normal vector to to satisfy that the triple of the orientation of and coincides with the right-handed orientation of , and thus, we represent the orientation of .
2.2. Biquandles, tribrackets, local biquandles
Definition 2.1**.**
([4, 9]) A biquandle is a set equipped with binary operations satisfying the following axioms.
- •
For any , .
- •
For any , the map sending to is bijective.
- For any , the map sending to is bijective.
- The map defined by is bijective.
- •
For any ,
[TABLE]
We denote it by or by for short unless it causes confusion. For , we denote by and the elements and , respectively. A biquandle with is called a quandle [7, 11], which is also denoted by .
Definition 2.2**.**
Let be a biquandle.
- (1)
A -set is a set equipped with a map satisfying the following axiom:
For any , sending to is bijective.
For any and , .
We denote it by or by for short unless it causes confusion. We denote by the element for and .
- (2)
A -set is strongly connected if it satisfies the following axiom:
For any , the map sending to is bijective.
We denote by the element for .
Definition 2.3**.**
A shadow biquandle is a pair of a biquandle and a -set . We denote it by or by for short unless it causes confusion. In particular, when is a quandle, the shadow biquandle is also called a shadow quandle and denoted by .
Lemma 2.4**.**
Let be a shadow biquandle. For any and , we have
- (1)
\big{(}x*^{-1}(a\mathbin{\underline{*}}b)\big{)}*^{-1}b=\big{(}x*^{-1}(b\mathbin{\overline{*}}a)\big{)}*^{-1}a,
- (2)
(x*^{-1}a)*(b\mathbin{\overline{*}}^{-1}a)=(x*b)*^{-1}\big{(}a\mathbin{\underline{*}}(b\mathbin{\overline{*}}^{-1}a)\big{)},
- (3)
(x*^{-1}b)*(a\mathbin{\underline{*}}^{-1}b)=(x*a)*^{-1}\big{(}b\mathbin{\overline{*}}(a\mathbin{\underline{*}}^{-1}b)\big{)}.
Proof.
We leave the proof of this lemme to the reader, refer also to Figure 2.
∎
Lemma 2.5**.**
Let be a shadow biquandle such that is strongly connected. For and , we have
- (1)
,
- (2)
,
- (3)
,
- (4)
.
Proof.
This lemma follows from the above definitions, refer also to Figure 3.
∎
Lemma 2.6**.**
Let be a latin quandle, that is, it satisfies that
- •
for any , is bijective.
Let and let be defined by . Then is a strongly connected -set.
Proof.
This can be easily shown by direct observation. We leave the proof of this lemma to the reader. ∎
In this paper, for a positive integer , means the quotient ring , and means the Laurent polynomial ring with coefficients in .
Example 2.7**.**
For a positive integer , let . We define by , and then, is a quandle called the dihedral quandle of order . Let . Then is an -set with defined by .
In particular when is an odd number other than , since is latin, is strongly connected by Lemma 2.6. We then have for .
The next example is a generalization of Example 2.7.
Example 2.8**.**
For a positive integer and an ideal of , let be the quotient ring. We define by , and then, is a quandle called an Alexander quandle. Let . Then is a -set with defined by .
In particular when is a unit in , since is latin, is strongly connected by Lemma 2.6. We then have for .
Definition 2.9**.**
(cf. [15, 17]) A knot-theoretic horizontal-ternary-quasigroup is a pair of a set and a ternary operation satisfying the following property:
- (1)
- (i)
For any , there exists a unique such that ,
- (ii)
For any , there exists a unique such that ,
- (iii)
For any , there exists a unique such that . 2. (2) For any , it holds that
[TABLE]
We call the operation a horizontal-tribracket.
Definition 2.10**.**
([17]) Let be a knot-theoretic horizontal-ternary-quasigroup. For each , we define two operations by
[TABLE]
We call the local biquandle associated with . In this paper, for simplicity, we often omit the subscript by as , , , and unless it causes confusion.
The next two examples are related to Examples 2.7 and 2.8, respectively, which will be shown in Subsection 5.2.
Example 2.11**.**
For a positive integer , let . We define a map by
[TABLE]
and then, is a horizontal-tribracket. We call it the dihedral horizontal-tribracket of order .
The local biquandle associated with this has the operations defined by
[TABLE]
Example 2.12**.**
For a positive integer and an ideal of , let be the quotient ring. We define a map by
[TABLE]
and then, is a horizontal-tribracket. We call it an Alexander horizontal-tribracket.
The local biquandle associated with this has the operations defined by
[TABLE]
3. Local biquandle homology groups/cocycle invariants and shadow biquandle homology groups/cocycle invariants
3.1. Shadow biquandle homology groups
Let be a shadow biquandle.
Let be the free -module generated by the elements of if , and otherwise. We define a homomorphism by
[TABLE]
if , and otherwise. Then is a chain complex. Let be a submodule of that is generated by the elements of
[TABLE]
Then is a subchain complex of . Therefore the chain complex
[TABLE]
is induced. We call the homology group of the th shadow biquandle homology group of .
For an abelian group , we define the chain and cochain complexes by
[TABLE]
Let and . The nth homology group and nth cohomology group of with coefficient group are defined by
[TABLE]
The th cocycle group with coefficient group is denoted by . Note that we omit the coefficient group if as usual.
3.2. Shadow biquandle colorings of link diagrams and cocycle invariants
Let be a shadow biquandle. Let be a diagram of a link .
Definition 3.1**.**
A * (biquandle) -coloring* of is a map satisfying the following condition:
- •
For a crossing composed of under-semi-arcs and over-semi-arcs as depicted in Figure 4,
- –
, and
- –
hold, see also Figure 5.
Definition 3.2**.**
A * (shadow biquandle) -coloring* of is a map satisfying the following condition:
- •
The restriction is a -coloring of .
- •
.
- •
For a semi-arc whose normal vector points from a region to a region as depicted in Figure 4, holds, see also Figure 5.
We denote by the set of -colorings of .
Proposition 3.3**.**
(cf. [8])* Let and be connected diagrams of links. If and represent the same link, then there exists a bijection between and .*
Next, we show how to obtain a cocycle invariant by using the -colorings of a diagram.
Let be a -coloring of . We define the local chain at each crossing by
[TABLE]
when , and , where , and are the region, under-semi-arc and over-semi-arc of as depicted in Figure 4, see also Figure 5.. We define a chain by
[TABLE]
Let be an abelian group. For a -cocycle , we define
[TABLE]
as multisets. Then we have the following theorem:
Theorem 3.4**.**
(cf. [8])* and are invariants of .*
3.3. Shadow biquandle colorings of surface-link diagrams, cocycle invariants
Let be a shadow biquandle. Let be a diagram of a surface-link .
Definition 3.5**.**
A * (biquandle) -coloring* of is a map satisfying the following condition:
- •
For a double point curve composed of under-semi-sheets and over-semi-sheets as depicted in Figure 6,
- –
, and
- –
hold, see also Figure 7.
Definition 3.6**.**
A * (shadow biquandle) -coloring* of is a map satisfying the following condition:
- •
The restriction is a -coloring of .
- •
.
- •
For a semi-sheet whose normal vector points from a region to a region as depicted in Figure 6, holds, see also Figure 7.
We denote by the set of -colorings of .
Proposition 3.7**.**
(cf. [8])* Let and be connected diagrams of surface-links. If and represent the same surface-link, then there exists a bijection between and .*
Next, we show how to obtain a cocycle invariant by using the -colorings of a diagram.
Let be a -coloring of . We define the local chain at each triple point by
[TABLE]
when , , and , where , , and are the region, bottom-semi-sheet, middle-semi-sheet and top-semi-sheet of as depicted in Figure 8, see also Figure 9. We define a chain by
[TABLE]
Let be an abelian group. For a -cocycle , we define
[TABLE]
as multisets. Then we have the following theorem:
Theorem 3.8**.**
(cf. [8])* and are invariants of .*
3.4. Local biquandle homology groups
Let be a knot-theoretic horizontal-ternary-quasigroup and the local biquandle associated with . Let . Let be the free -module generated by the elements of
[TABLE]
if , and otherwise. We define a homomorphism by
[TABLE]
if , and otherwise. Then is a chain complex. Let be a submodule of that is generated by the elements of
[TABLE]
Then is a subchain complex of . Therefore the chain complex
[TABLE]
is induced. We call the homology group of the th local biquandle homology group of .
For an abelian group , we define the chain and cochain complexes by
[TABLE]
Let and . The nth homology group and nth cohomology group of with coefficient group are defined by
[TABLE]
The th cocycle group with coefficient group is denoted by . Note that we omit the coefficient group if as usual.
3.5. Local biquandle colorings of link diagrams, cocycle invariants
Let be a knot-theoretic horizontal-ternary-quasigroup and the local biquandle associated with . Let be a connected diagram of a link .
Definition 3.9**.**
A (local biquandle) -coloring of is a map satisfying the following condition:
- •
For a crossing composed of under-semi-arcs and over-semi-arcs as depicted in Figure 4, let . Then
- –
,
- –
, and
- –
hold, where , see also Figure 10.
We denote by the set of -colorings of .
Proposition 3.10**.**
([17])* Let and be connected diagrams of links. If and represent the same link, then there exists a bijection between and .*
Next, we show how to obtain a cocycle invariant by using the -colorings of a connected diagram.
Let be an -coloring of . We define the local chain at each crossing by
[TABLE]
when and , where and are the under-semi-arc and over-semi-arc of as depicted in Figure 4, see also Figure 10. We define a chain by
[TABLE]
Let be an abelian group. For a -cocycle , we define
[TABLE]
as multisets. Then we have the following theorem:
Theorem 3.11**.**
([17])* and are invariants of .*
3.6. Local biquandle colorings of surface-link diagrams, cocycle invariants
Let be a knot-theoretic horizontal-ternary-quasigroup and the local biquandle associated with . Let be a connected diagram of a surface-link .
Definition 3.12**.**
A (local biquandle) -coloring of is a map satisfying the following condition:
- •
For a double point curve composed of under-semi-sheets and over-semi-sheets as depicted in Figure 6, let . Then
- –
,
- –
, and
- –
hold, where , see also Figure 11.
We denote by the set of -colorings of .
Proposition 3.13**.**
([17])* Let and be connected diagrams of surface-links. If and represent the same surface-link, then there exists a bijection between and .*
Next, we show how to obtain a cocycle invariant by using the -colorings of a connected diagram.
Let be an -coloring of . We define the local chain at each triple point by
[TABLE]
when , and , where , and are the bottom-semi-sheet, middle-semi-sheet and top-semi-sheet of as depicted in Figure 8, see also Figure 12. We define a chain by
[TABLE]
Let be an abelian group. For a -cocycle , we define
[TABLE]
as multisets. Then we have the following theorem:
Theorem 3.14**.**
([17])* and are invariants of .*
4. Main results
4.1. Corresponding tribrackets and local biquandles
Theorem 4.1**.**
Given a shadow biquandle such that is strongly connected, we have a horizontal-tribracket defined by
[TABLE]
Proof.
We first show the second equality. It holds since
[TABLE]
Next we show the first equality by checking the horizontal-tribracket axioms one by one.
()-(i) Suppose that are given. Let , where . We then have
[TABLE]
The uniqueness of the above holds as follows: Assume that for some . We then have
[TABLE]
Hence we have
[TABLE]
Then we have
[TABLE]
()-(ii) Suppose that are given. Let , where . We then have
[TABLE]
The uniqueness of the above holds as follows: Assume that for some . We then have
[TABLE]
Hence we have
[TABLE]
Therefore we have
[TABLE]
()-(iii) Suppose that are given. Let , where such that for the bijection in Definition 2.1, and where it holds that
[TABLE]
We then have
[TABLE]
The uniqueness of the above holds as follows: Assume that for some . We then have
[TABLE]
and
[TABLE]
Hence we have
[TABLE]
and
[TABLE]
Since there exists a unique element such that by Definition 2.1, we have
[TABLE]
Therefore we have
[TABLE]
() For , we have
[TABLE]
and
[TABLE]
This completes the proof. ∎
Definition 4.2**.**
For a shadow biquandle such that is strongly connected, we call the horizontal-tribacket given in Theorem 4.1 the corresponding horizontal-tribacket of . We call the local biquandle associated with the corresponding horizontal-tribacket of the corresponding local biquandle of .
4.2. Correspondence between (co)homology groups
Let be a shadow biquandle such that is strongly connected. Let be the corresponding horizontal-tribracket of , that is, it is defined by
[TABLE]
Let be the corresponding local biquandle of , that is, it is the local biquandle associated with the above . Define a homomorphism by
[TABLE]
if , and otherwise.
Lemma 4.3**.**
* is a bijective chain map.*
Proof.
It is sufficient to consider the cases that .
We first show that is well-defined. For , suppose that for some . We then have
[TABLE]
This implies is well-defined since \mu_{n}\big{(}D_{n}^{\rm sb}(B,X)\big{)}\subset D_{n}^{\rm lb}(X) when we regard as a homomorphism from to .
Next we show that is bijective. Define a homomorphism by
[TABLE]
if , and otherwise. Then if for some ,
[TABLE]
This implies that is well-defined since \eta_{n}\big{(}D_{n}^{\rm lb}(X)\big{)}\subset D_{n}^{\rm sb}(B,X) when we regard as a homomorphism from to . For , we have
[TABLE]
and
[TABLE]
Hence is the inverse map of , and thus, is bijective.
Lastly, we show that is a chain map. We have
[TABLE]
and
[TABLE]
We can easily see that the terms (1) coincide with the terms (3). The terms (2) coincide with the terms (4) because for , it holds that
[TABLE]
and for , it holds that
[TABLE]
Therefore we have
[TABLE]
and thus, is a chain map.
This completes the proof. ∎
The bijective chain map induces an isomorphism defined by
[TABLE]
if , and otherwise.
Moreover, for an abelian group , the bijective chain map induces the bijective chain map , and hence, we have an isomorphism . The bijective cochain map induces the bijective cochain map defined by , and hence, we have an isomorphism . Thus we have the following theorem:
Theorem 4.4**.**
Let be a shadow biquandle such that is strongly connected, and be the corresponding local biquandle of . Let be an abelian group. Then for any , we have
[TABLE]
4.3. Correspondence between cocycle invariants of links
Let be a shadow biquandle such that is strongly connected. Let be the corresponding horizontal-tribracket of , that is, it is defined by
[TABLE]
Let be the corresponding local biquandle of , that is, it is the local biquandle associated with the above .
Let be a connected diagram of a link .
Lemma 4.5**.**
There exists a bijection .
Proof.
We set a map as follows: Let . For a semi-arc whose normal vector points from a region to a region as shown in the right of Figure 4, we assign to the semi-arc , where and , see also Figure 13. Then the assignment determines an -coloring . Indeed, since
[TABLE]
for a crossing of as shown in the left of Figure 14, for the same crossing of as shown in the right of Figure 14, the conditions of a local biquandle coloring in Definition 3.12 hold as follows:
[TABLE]
The inverse map is defined as follows: Let . For a semi-arc whose normal vector points from a region to a region as shown in the right of Figure 4, we assign to the region , to the region , and to the semi-arc , where , see also Figure 13. Then the assignment determines a -coloring . Indeed, since for a crossing of as shown in the right of Figure 14, for the same crossing of as shown in the left of Figure 14, the conditions of a shadow biquandle coloring in Definition 3.6 hold as follows:
[TABLE]
For a semi-arc of as shown in the right of Figure 4,
[TABLE]
and thus, the condition of a shadow biquandle coloring around each semi-arc also holds.
Therefore is bijective.
∎
We continue to use the bijection .
Let be the bijective chain map defined in Subsection 4.2, that is, it is defined by
[TABLE]
We note that the inverse map of is defined by
[TABLE]
Let and such that . At a crossing of as depicted in Figure 4, we have
[TABLE]
where and , see also Figure 14. This implies that W^{\rm LB}(D,C^{\prime})=\mu_{2}\big{(}W^{\rm SB}(D,C)\big{)}. Thus we have
[TABLE]
We note that since is an isomorphism, \mathcal{H}^{\rm SB}(D)=(\mu_{2}^{\ast})^{-1}\big{(}\mathcal{H}^{\rm LB}(D)\big{)} holds. This implies that as link invariants, and are the same.
Let be an abelian group. Let and such that . We then have
[TABLE]
for each crossing , and thus, \theta\big{(}W^{\rm SB}(D,C)\big{)}=\theta^{\prime}(W^{\rm LB}(D,C^{\prime})) holds. This implies that .
As a consequence, we have the following theorem:
Theorem 4.6**.**
Let be a link. Let be a shadow biquandle such that is strongly connected, and be the corresponding local biquandle of . Then we have
[TABLE]
Moreover for an abelian group , let and such that . Then we have
[TABLE]
4.4. Correspondence between cocycle invariants of surface-links
Let be a shadow biquandle such that is strongly connected. Let be the corresponding horizontal-tribracket of , that is, it is defined by
[TABLE]
Let be the corresponding local biquandle of , that is, it is the local biquandle associated with the above .
Let be a connected diagram of a surface-link .
Lemma 4.7**.**
There exists a bijection .
Proof.
Here, we show only how to construct a bijection , and the details are left to the reader, refer to the proof of Lemma 4.5.
We set a map as follows: Let . For a semi-sheet whose normal vector points from a region to a region as shown in the right of Figure 6, we assign to the semi-sheet , where and , see also Figure 15. Then the assignment determines an -coloring .
The inverse map is defined as follows: Let . For a semi-sheet whose normal vector points from a region to a region as shown in the right of Figure 6, we assign to the region , to the region , and to the semi-sheet , where , see also Figure 15. Then the assignment determines a -coloring .
Therefore is bijective.
∎
We continue to use the bijection .
Let be the bijective chain map defined in Subsection 4.2, that is, it is defined by
[TABLE]
We note that the inverse map of is defined by
[TABLE]
Let and such that . At a triple point of as depicted in Figure 8, we have
[TABLE]
where , and , see also Figure 16. This implies that W^{\rm LB}(D,C^{\prime})=\mu_{3}\big{(}W^{\rm SB}(D,C)\big{)}. Thus we have
[TABLE]
We note that since is an isomorphism, \mathcal{H}^{\rm SB}(D)=(\mu_{3}^{\ast})^{-1}\big{(}\mathcal{H}^{\rm LB}(D)\big{)} holds. This implies that as surface-link invariants, and are the same.
Let be an abelian group. Let and such that . We then have
[TABLE]
for each triplepoint , and thus, \theta\big{(}W^{\rm SB}(D,C)\big{)}=\theta^{\prime}(W^{\rm LB}(D,C^{\prime})) holds. This implies that .
As a consequence, we have the following theorem:
Theorem 4.8**.**
Let be a surface-link. Let be a shadow biquandle such that is strongly connected, and be the corresponding local biquandle of . Then we have
[TABLE]
Moreover for an abelian group , let and such that . Then we have
[TABLE]
5. Remarks
5.1. Shadow biquandle theory and Niebrzydowski’s theory
In [14, 15, 16], region colorings of link diagrams by using algebraic structures called knot-theoretic ternary quasigroups were studied and used to define invariants of links and surface-links. Furthermore, Niebrzydowski in [15, 16] introduced a (co)homology theory of the algebraic structures, and defined a cocycle invariant for links and surface-links. In this subsection, we denote by and the th Niebrzydowski’s homology group and cohomology group, respectively, for a given knot-theoretic horizontal-ternary-quasigroup and an abelian group . Note that several versions of Niebrzydowski’s (co)homology groups were defined in [15, 16], and in this subsection, his (co)homology groups mean the (co)homology groups reviewed in [17]. In addition, we denote by and the link invariants for a link using the homology group and a -cocycle of his homology theory, respectively. We denote by and the surface-link invariants for a surface-link using the homology group and a -cocycle of his homology theory, respectively, see [17] for details.
In [17], we introduced local biquandle theory to show that the Niebrzydowski’s (co)homology theory can be interpreted as local biquandle (co)homology theory. On other words, the Niebrzydowski’s (co)homology theory can be interpreted similarly as biquandle (co)homology theory since local biquandle (co)homology theory is an analogy of biquandle (co)homology theory. Moreover through an isomorphism between two cohomology groups, we showed that Niebrzydowski’s cocycle invariants and local biquandle cocycle invariants are the same.
Considering the main results shown in Section 4 in this paper together with the results shown in [17], we have the following corollaries:
Corollary 5.1**.**
Let be a shadow biquandle such that is strongly connected, the corresponding horizontal-tribracket of , and the corresponding local biquandle of . Let be an abelian group. Then for any , we have
[TABLE]
and
[TABLE]
Corollary 5.2**.**
Let be a link. Let be a shadow biquandle such that is strongly connected, the corresponding horizontal-tribracket of , and the corresponding local biquandle of . Then
[TABLE]
are the same as link invariants. Moreover for an abelian group , let , and such that and , where is the first cocycle group of the Niebrzydowski’s (co)homology theory and is the bijective chain map defined in [17]. Then we have
[TABLE]
Corollary 5.3**.**
Let be a surface-link. Let be a shadow biquandle such that is strongly connected, the corresponding horizontal-tribracket of , and the corresponding local biquandle of . Then
[TABLE]
are the same as link invariants. Moreover for an abelian group , let , and such that and , where is the second cocycle group of the Niebrzydowski’s (co)homology theory and is the bijective chain map defined in [17]. Then we have
[TABLE]
5.2. Examples and Mochizuki’s cocycles
Example 5.4**.**
For a positive integer and an ideal of , let is the shadow quandle, with and , defined in Example 2.8. Suppose that is a unit in . Since is strongly connected and we have
[TABLE]
it holds that
[TABLE]
by Theorem 4.1, where we note that . Thus Example 2.12 is related to Example 2.8.
Example 5.5**.**
Let be an odd prime number. Let be the shadow quandle, with and , defined in Example 2.7. Since is strongly conneced and we have
[TABLE]
it holds that
[TABLE]
by Theorem 4.1, where we note that . Thus Example 2.11 is related to Example 2.7.
The shadow (bi)quandle -cocycle defined by
[TABLE]
is called a Mochizuki’s cocycle. Let . The corresponding local biquandle -cocycle , i.e. such that , is defined by
[TABLE]
since in by Fermat’s little theorem, where the numerator is calculated in and it is divisible by . We may define by
[TABLE]
Furthermore, the corresponding Niebrzydowski’s -cocycle , i.e. such that , is defined by
[TABLE]
where is the bijective chain map defined in [17], is the corresponding vertical-tribracket defined in [17] and for this case, and where the numerator is calculated in and it is divisible by . We may define by
[TABLE]
The Mochizuki’s cocycle induces a shadow biquadle -cocycle by
[TABLE]
The corresponding local biquandle -cocycle , i.e. such that , is defined by
[TABLE]
where the numerator is calculated in and it is divisible by . We may define by
[TABLE]
Furthermore, the corresponding Niebrzydowski’s -cocycle , i.e. such that , is defined by
[TABLE]
where is the bijective chain map defined in [17], is the corresponding vertical-tribracket defined in [17] and for this case, and where the numerator is calculated in and it is divisible by . We may define by
[TABLE]
Acknowledgments
The author wishes to express her thanks to Natsumi Oyamaguchi for several helpful comments.
The author was supported by JSPS KAKENHI Grant Number 16K17600.
The reference list from the paper itself. Each links out to its DOI / PubMed record.
- 1[1] S. Carter, M. Elhamdadi and M. Saito, Homology theory for the set-theoretic Yang-Baxter equation and knot invariants from generalizations of quandles , Fund. Math. 184 (2004), 31–54.
- 2[2] J. S. Carter, S. Kamada, and M. Saito, Geometric interpretations of quandle homology , J. Knot Theory Ramifications 10 (2001), 345–386.
- 3[3] W. Choi, D. Needell and S. Nelson, Boltzmann enhancements of biquasile counting invariants , ar Xiv:1704.02555.
- 4[4] R. Fenn, C. Rourke and B. Sanderson, An introduction to species and the rack space , Topics in knot theory (Erzurum, 1992), 33–55, NATO Adv. Sci. Inst. Ser. C Math. Phys. Sci. 399 , Kluwer Acad. Publ., Dordrecht, 1993.
- 5[5] R. Fenn, C. Rourke and B. Sanderson, James bundles and applications , preprint found at http://www.maths.warwick.ac.uk/ bjs/.
- 6[6] R. Fenn, M. Jordan-Santana and L. Kauffman, Biquandles and virtual links , Topology Appl. 145 (2004), no. 1–3, 157–175.
- 7[7] D. Joyce, A classifying invariant of knots, the knot quandle , J. Pure Appl. Alg. 23 (1982), 37–65.
- 8[8] S. Kamada, A. Kawauchi, J. Kim and S. Y. Lee, Biquandle cohomology and state-sum invariants of links and surface-links , ar Xiv:1803.03137.
