Metric Topologies on Multiset Spaces as Topological Monoids and Their Group Completion

Basic Information

• Paper ID: 2510.10080
• Title: Metric Topologies on Multiset Spaces as Topological Monoids and Their Group Completion
• Author: Donghan Kim (KAIST, South Korea)
• Classification: math.MG (Metric Geometry)
• Publication Date: October 11, 2025
• Paper Link: https://arxiv.org/abs/2510.10080

Abstract

This paper constructs the multiset space $\mathbb{N}[X]$ over a metric space $X$, which simultaneously enjoys ideal topological properties and admits a natural matching metric $d_{\mathbb{N}[X]}$, making it a metrizable abelian topological monoid whose structure is compatible with the original metric on $X$. This framework naturally extends to the free abelian group $\mathbb{Z}[X]$, where the metric $d_{\mathbb{Z}[X]}$ induces a metrizable abelian topological group structure. We further determine the metric completion of $\mathbb{N}[X]$, proving that it carries a canonical extension of the matching metric.

Research Background and Motivation

Problem Context

1. Importance of Multisets: Multisets are collections of objects that allow repetitions and have important applications in algebra, combinatorics, and computer science, including database semantics, query languages, and data analysis.
2. Topological Structure Problem: When the underlying space $X$ is a topological space, how to endow its multiset space with a compatible topological structure is a natural question. Classical constructions include the Dold-Thom infinite symmetric product and McCord's classifying space.
3. Metrizability Problem: Although classical constructions can endow multiset spaces with abelian topological monoid structures when $X$ is compactly generated, even when $X$ is metrizable, the infinite symmetric product is typically not metrizable.

Core Problem

Research Motivation: Given a metric space $(X,d)$, can one construct a metric on the associated multiset space such that:

• It admits an isometric embedding of $X$ (compatible with the original metric)
• It makes the addition operation continuous, thus forming an abelian topological monoid

Core Contributions

1. Construction of Metrizable Multiset Spaces: For a metric space $(X,d)$, we construct the multiset space $\mathbb{N}[X]$ equipped with the matching distance $d_{\mathbb{N}[X]}$, making it a metrizable topological abelian monoid.
2. Establishment of Topological Equivalence Conditions: We prove that when the basepoint $e$ is isolated, $\mathbb{N}[X]$ coincides with the infinite symmetric product $SP(X)$; when $e$ is not isolated, $SP(X)$ may not be first countable, and hence not metrizable.
3. Extension to Free Abelian Groups: We extend the construction to the free abelian group $\mathbb{Z}[X]$, defining the metric $d_{\mathbb{Z}[X]}$ such that $\mathbb{Z}[X]$ becomes a metrizable abelian topological group.
4. Isometric Embedding Sequences: We prove that the canonical maps $X \hookrightarrow \mathbb{N}[X] \hookrightarrow \mathbb{Z}[X]$ form a sequence of isometric embeddings.
5. Completion Characterization: We establish completeness criteria, characterizing the completion of $\mathbb{N}[X]$ as the space $\overline{\mathbb{N}[X]}$ equipped with the extended matching metric $d_\ell$.

Methodology Details

Task Definition

Input: Metric space with basepoint $(X,e,d)$
Output: Metrizable multiset space $\mathbb{N}[X]$ and free abelian group $\mathbb{Z}[X]$, both equipped with compatible metric structures

Core Construction

1. Matching Distance on Multiset Spaces

For elements $[x_1,\ldots,x_m]$ and $[y_1,\ldots,y_n]$ in $SP(X)$, the matching distance is defined as:

Definition 3.1: Choose $N \geq m+n$ and form padded lists: $x_N := [x_1,\ldots,x_m,\underbrace{e,\ldots,e}_{N-m}], \quad y_N := [y_1,\ldots,y_n,\underbrace{e,\ldots,e}_{N-n}]$

The matching distance is defined as: $d_{\mathbb{N}[X]}([x_1,\ldots,x_m],[y_1,\ldots,y_n]) := \inf_{N\geq m+n} \min_{\sigma \in S_N} \sum_{i=1}^N d(x_N(i), y_N(\sigma(i)))$

2. Metric on Free Abelian Groups

For elements in $\mathbb{Z}[X]$, we use the positive and negative part decomposition:

Definition 5.1: For $x,y \in \mathbb{Z}[X]$, let $x = x^+ - x^-$ and $y = y^+ - y^-$. Define: $d_{\mathbb{Z}[X]}(x,y) := d_{\mathbb{N}[X]}(x^+ + y^-, y^+ + x^-)$

Technical Innovations

1. Well-definedness of the Metric: We prove by induction that the matching distance is independent of the choice of repetitions of the basepoint $e$.
2. Topological Monoid Structure: We prove that the addition operation is 1-Lipschitz continuous, hence continuous.
3. Isometric Embeddings: We establish the isometric embedding chain $SP^n(X) \hookrightarrow \mathbb{N}[X] \hookrightarrow \mathbb{Z}[X]$.

Theoretical Results

Main Theorems

Theorem 3.4 (Abelian Topological Monoid Structure): The metric $d_{\mathbb{N}[X]}$ induces an abelian topological monoid structure on $\mathbb{N}[X]$.

Theorem 3.8 (Topological Equivalence): Assume the basepoint $e \in X$ is isolated. Then the topology induced by the metric $d_{\mathbb{N}[X]}$ on $SP(X)$ coincides with the direct limit topology.

Theorem 5.3 (Abelian Topological Group Structure): The metric $d_{\mathbb{Z}[X]}$ induces an abelian topological group structure on $\mathbb{Z}[X]$.

Theorem 4.10 (Metric Completion): If $(X,d)$ is complete, then the metric completion of $(\mathbb{N}[X], d_{\mathbb{N}[X]})$ is isometric to $(\overline{\mathbb{N}[X]}, d_\ell)$.

Topological Properties

Proposition 4.1:

• If $X$ is connected, then $\mathbb{N}[X]$ is connected
• If $X$ is path-connected, then $\mathbb{N}[X]$ is path-connected

Counterexample Analysis

Example 3.11: Let $X = [0,1]$ with basepoint $e = 0$. Consider the set $K := \{[1], [1/2], [1/3], \ldots\}$. In the direct limit topology, $K$ is closed, but in the $d_{\mathbb{N}[X]}$-metric, the sequence $([1/n])$ converges to $[e]$, showing that the metric topology is strictly coarser than the direct limit topology.

Completion Theory

ℓ¹-Multiset Spaces

Definition 4.5: Define an equivalence relation, and $\overline{\mathbb{N}[X]}$ as the quotient of the sequence space $X^*$, where: $X^* = \{x = (x_i)_{i \in \mathbb{N}} : \sum_{i=1}^{\infty} d(x_i, e) < \infty\}$

The extended matching distance is: $d_\ell([x], [y]) := \inf_{\sigma \in S_{\infty}} \sum_{i=1}^{\infty} d(x_i, y_{\sigma(i)})$

Proposition 4.8: If $(X,d)$ is complete, then $(\overline{\mathbb{N}[X]}, d_\ell)$ is complete.

Related Work

1. Classical Constructions: The Dold-Thom infinite symmetric product and McCord's classifying space provide topological structures on multiset spaces, but are typically not metrizable.
2. Metric Monoids: Recent work has studied metric monoids of multisets in Banach and combinatorial settings.
3. Matching Metrics: Multisets are commonly compared using matching or earth-mover type metrics.

Conclusions and Discussion

Main Conclusions

This paper successfully resolves the metrizability problem for multiset spaces over metric spaces, constructing metric structures that preserve compatibility with the original metric while maintaining good topological properties.

Theoretical Significance

1. Bridging Algebra and Topology: Naturally combining algebraic structures (monoids, groups) with metric geometry.
2. Completion Theory: Providing explicit characterization of metric completions of multiset spaces.
3. Topological Equivalence Conditions: Clarifying when the new construction coincides with classical constructions.

Limitations

1. Completeness: $\mathbb{N}[X]$ may not be complete even when $X$ is complete.
2. Basepoint Dependence: The construction depends on the choice of basepoint.
3. Computational Complexity: Computing the matching distance involves combinatorial optimization problems.

Future Directions

1. Investigate generalizations to more general classes of metric spaces
2. Explore efficient algorithms for computing matching distances
3. Apply to concrete problems in mathematics and computer science

In-Depth Evaluation

Strengths

1. Theoretical Completeness: Constructs a complete theoretical framework from basic definitions to completion theory.
2. Technical Rigor: Detailed proofs, particularly in verifying metric axioms and establishing topological equivalence.
3. Novelty: First systematic metrization theory for multisets over metric spaces.
4. Application Potential: Provides new mathematical tools for database theory, combinatorial optimization, and related fields.

Weaknesses

1. Practical Applications: Lacks demonstration of concrete application examples.
2. Computational Aspects: Insufficient discussion of computational complexity of matching distance.
3. Generalizability: Limited discussion of more general settings (e.g., pseudometric spaces).

Impact

This is a high-quality pure mathematics paper making significant contributions to the intersection of metric geometry and algebraic topology, with potential impact on related applied mathematics fields.

Applicable Scenarios

1. Theoretical Mathematics: Research in metric geometry and algebraic topology
2. Computer Science: Database theory, algorithm design
3. Applied Mathematics: Optimization theory, data analysis

References

The paper cites 22 relevant references spanning multiple fields including topology, combinatorics, and computer science, providing a solid theoretical foundation for the research.