Finite Type Link Homotopy Invariants

Blake Mellor

DOI: https://doi.org/10.48550/arXiv.math/9807162

1999-12-08

Abstract:Bar-Natan used Chinese characters to show that finite type invariants classify string links up to homotopy. In this paper, I construct the correct spaces of chord diagrams and Chinese characters for links up to homotopy. I use these spaces to show that the only rational finite type invariants of link homotopy are the pairwise linking numbers of the components.

Geometric Topology

What problem does this paper attempt to address?

The problem that this paper attempts to solve is about the finite - type invariants of link homotopy. Specifically, the author Blake Mellor tries to construct the correct spaces of chord diagrams and unitrivalent diagrams applicable to link homotopy, and uses these spaces to prove that for link homotopy, the only rational finite - type invariants are the linking numbers between components.

### Specific background of the problem

1. **Definition of finite - type invariants**:
- Finite - type invariants are introduced by studying certain groups related to the cohomology of the space of knots. Vassiliev first proposed the concept of Vassiliev or finite - type knot invariants in 1990.
- Bar - Natan et al. further extended these concepts to chord diagrams and unitrivalent diagrams and showed how to represent and calculate finite - type invariants using these diagrams.

2. **Link homotopy**:
- Link homotopy is a concept introduced by Milnor, which means that two links can be transformed into each other through a series of ambient isotopies and crossing changes within the same component, but crossing changes between different components are not allowed.

3. **Existing results**:
- Bar - Natan has proved that finite - type invariants can classify string links up to homotopy.
- However, for general links, the situation is more complicated because the operation of closing multiple components is non - trivial and requires the introduction of additional relations.

### Main contributions of the paper

- **Constructing appropriate diagram spaces**:
- The author constructs the spaces of chord diagrams and unitrivalent diagrams applicable to link homotopy and defines the relations in these spaces (such as (*) relations).

- **Proving uniqueness**:
- Through detailed mathematical derivations, the author proves that for link homotopy, the only rational finite - type invariants are the linking numbers between components. This means that in the sense of link homotopy, other complex invariants do not provide new information except for the linking numbers.

### Main conclusions

Finally, the paper draws the following important conclusion:

\[
\text{The only rational finite - type invariant under link homotopy is the linking number between components.}
\]

This shows that within the framework of link homotopy, the linking number is the most important invariant, and other more complex invariants (such as Milnor invariants) cannot be directly used as finite - type invariants due to their uncertainty and complexity.

### Related corollaries

- **Corollary 1**: The unitrivalent diagram space \( B_{hl}(k) \) of link homotopy is isomorphic to the polynomial algebra \( R[x_{ij}] \), where each \( x_{ij} \) is a term of degree 1 and \( 1\leq i < j\leq k \).
- **Corollary 2**: The linking number between components of a link is the only real - valued finite - type link - homotopy invariant.

These conclusions provide an important theoretical basis for understanding link homotopy and point out the direction for future research.