Compact subspaces of the space of separately continuous functions with the cross-uniform topology

Basic Information

• Paper ID: 2406.05705
• Title: Compact subspaces of the space of separately continuous functions with the cross-uniform topology
• Authors: Oleksandr Maslyuchenko, Vadym Myronyk, Roman Ivasiuk
• Classification: math.GN (General Topology)
• Publication Date: June 11, 2024
• Paper Link: https://arxiv.org/abs/2406.05705

Abstract

This paper investigates two natural topologies on the space $S(X\times Y,Z)$ of all separately continuous functions defined on the product of two topological spaces $X$ and $Y$ and taking values in a topological or metric space $Z$: the cross-open topology and the cross-uniform topology. The authors prove that these two topologies coincide when $X$ and $Y$ are pseudocompact spaces and $Z$ is a metric space. The main result shows that for infinite compact spaces $X$, $Y$ and a metrizable space $Z$ containing $\mathbb{R}$, a compact space $K$ embeds into $S(X\times Y,Z)$ if and only if the weight of $K$ is less than the sharp cellularity of $X$ and $Y$.

Research Background and Motivation

1. Research Problem: This paper primarily investigates the topological structure of separately continuous function spaces, particularly the characterization of compact subspaces. Separately continuous functions are functions that are continuous in each variable when the other variables are held fixed.
2. Problem Significance: Separately continuous functions occupy an important position in multivariable function theory. The topological properties of their function spaces are crucial for understanding function convergence and continuity. The characterization of compact subspaces helps elucidate the overall structure of these function spaces.
3. Limitations of Existing Research: Although it is known that compact subspaces of Baire class one function spaces $B_1(X)$ are Rosenthal compact, our understanding of the structure of compact subspaces in separately continuous function spaces remains limited.
4. Research Motivation: The authors expected to discover Rosenthal-type compact spaces in separately continuous function spaces, but found that the actual topological structure is simpler, which motivated a complete characterization of compact subspaces.

Core Contributions

1. Topological Equivalence Theorem: Proves that the cross-open topology and cross-uniform topology coincide when $X$ and $Y$ are pseudocompact spaces and $Z$ is a metric space.
2. Complete Characterization of Compact Embeddings: For infinite compact spaces $X$, $Y$ and a metrizable space $Z$ containing $\mathbb{R}$, completely characterizes which compact spaces can be embedded into $S(X\times Y,Z)$.
3. Weight Bound Theorem: Proves that a compact space $K$ embeds into $S(X\times Y,Z)$ if and only if $w(K) < \min\{c^♯(X), c^♯(Y)\}$.
4. Role of Eberlein Compacta: Establishes a reduction method from general compact spaces to Eberlein compact spaces, simplifying the analysis of the problem.

Detailed Methodology

Basic Definitions and Notation

For a product $P = X \times Y$ of topological spaces, the cross of a set $E \subseteq P$ is defined as: $\text{cr}E = (\text{pr}_X(E) \times Y) \cup (X \times \text{pr}_Y(E))$

where $\text{pr}_X$ and $\text{pr}_Y$ are coordinate projections.

Definitions of the Two Topologies

1. Cross-open Topology: Generated by the subbase $\{s \in S : s(A) \subseteq W\}$, where $A = G \cap C$, $C = \text{cr}\{p\}$, $G$ is an open set in $C$, and $W$ is an open set in $Z$.
2. Cross-uniform Topology: For a metric space $(Z,d)$, generated by the neighborhood subbase $\{t \in S : d(s(p),t(p)) < \varepsilon \text{ for all } p \in \text{cr}E\}$, where $E$ is a finite subset of $P$.

Key Technical Methods

1. Proof of Topological Equivalence

The proof of Proposition 2.1 employs a two-directional inclusion method:

• First proves that the cross-open topology is weaker than the cross-uniform topology, using pseudocompactness to guarantee the compactness of $s(K)$
• Then proves that the cross-uniform topology is weaker than the cross-open topology by constructing open neighborhoods via finite covers

2. Reduction to Eberlein Compacta

Lemma 4.3 provides a crucial reduction step:

• For any compact subspace $K \subseteq S(X \times Y, Z)$, constructs an Eberlein compact space $\tilde{X}$
• Proves that $K$ can be embedded into $S(\tilde{X} \times Y, Z)$, where $\tilde{X}$ is a continuous image of $X$
• This reduces the general case to the Eberlein compact case

3. Weight Estimation

By embedding the separately continuous function space into the continuous function space $C_p(X,M)$ (where $M = C(Y,Z)$), utilizing known weight estimation results: $w(K) \leq d(X) = c(X) < c(X)^+ = c^♯(X)$

Experimental Setup

This paper is purely theoretical research and does not involve numerical experiments. All results are derived through rigorous mathematical proofs.

Main Results

Theorem 6.2 (Necessity)

Statement: Let $X$, $Y$ be infinite compact spaces and $Z$ be a metrizable space. If $K$ is a compact subspace of $S(X\times Y,Z)$, then $w(K) < \min\{c^♯(X), c^♯(Y)\}$.

Proof Strategy:

1. Uses Lemma 4.3 to reduce to the Eberlein compact case
2. Constructs an embedding $\Phi: S \to C_p(X,M)$, where $M = C(Y,Z)$
3. Applies Propositions 6.1 and 5.2 to obtain weight estimates

Theorem 7.1 (Sufficiency)

Statement: Let $X$, $Y$ be infinite compact spaces and $Z$ be a metrizable space containing $\mathbb{R}$. A compact space $K$ embeds into $S(X\times Y,Z)$ if and only if $w(K) < \min\{c^♯(X), c^♯(Y)\}$.

Proof Strategy:

1. Necessity is proved by Theorem 6.2
2. Sufficiency is proved by constructing an explicit embedding:
   • Uses disjoint open set families to construct a function family $(f_i)_{i \in I}$
   • Defines a mapping $\Phi: [0,1]^I \to S(X \times Y)$
   • Proves that $\Phi$ is a continuous injection

Corollary 7.2

For infinite metrizable compact spaces $X$, $Y$, a compact space $K$ embeds into $S(X\times Y,Z)$ if and only if $K$ is metrizable.

Related Work

1. Baire Class One Functions: It is known that compact subspaces of Baire class one function spaces on Polish spaces are Rosenthal compact
2. Eberlein Compacta Theory: The Amir-Lindenstrauss theorem provides a structural characterization of Eberlein compact spaces
3. Function Space Topology: Generalizations of classical topologies such as pointwise topology and uniform topology
4. Cellularity Theory: Uses sharp cellularity $c^♯(X)$ as a key invariant

Conclusions and Discussion

Main Conclusions

1. Completely characterizes the structure of compact subspaces in separately continuous function spaces
2. Proves that these compact subspaces are simpler than expected (e.g., in the metrizable compact case, only metrizable compacta can be embedded)
3. Establishes a precise relationship between weight and sharp cellularity

Limitations

1. Results primarily address compact spaces; generalization to non-compact cases remains an open problem
2. Some proofs depend on the assumption that the space contains $\mathbb{R}$
3. The general theory of cross topologies requires further development

Future Directions

The paper poses three specific open problems:

1. Problem 1: Describe compact embeddings in the case of general topological spaces
2. Problem 2: Can Rosenthal compacta be embedded into separately continuous function spaces on certain Polish spaces?
3. Problem 3: Do there exist Polish spaces whose separately continuous function spaces contain non-metrizable compact subspaces?

In-Depth Evaluation

Strengths

1. Theoretical Completeness: Provides a complete solution to the compact embedding problem with precise and optimal results
2. Technical Innovation: Cleverly utilizes Eberlein compacta theory and the concept of sharp cellularity
3. Method Generality: The reduction technique and embedding construction methods have general value
4. Problem Formulation: Naturally generalizes classical function space theory

Weaknesses

1. Scope of Applicability: Main results are limited to the compact space case
2. Technical Dependencies: Some proofs depend on specific topological property assumptions
3. Practical Utility: As a purely theoretical result, practical applications are limited

Impact

1. Theoretical Contribution: Provides important structural theorems for separately continuous function space theory
2. Method Value: The reduction technique and construction methods may be applicable to other function space problems
3. Problem Inspiration: The posed open problems provide direction for subsequent research

Applicable Scenarios

1. Research on topological structures of function spaces
2. Embedding theory of compact spaces
3. Continuity theory of multivariable functions
4. Research on cardinal functions in general topology

References

The paper cites 17 important references, primarily including:

• Classical results by Amir-Lindenstrauss on Eberlein compacta
• Engelking's general topology textbook
• The authors' previous series of works on separately continuous function spaces
• Results by Benyamini-Rudin-Wage on weakly compact subsets of Banach spaces