On Hausdorff content maximal operator and Riesz potential for non-measurable functions

Background & Academic Lineage

The Origin & Academic Lineage

The problem addressed in this paper precisely originates from the fundamental limitations of classical integral theories when applied to functions that are not "well-behaved" in the traditional sense. Historically, the Riemann integral and later the more powerful Lebesgue integral were developed to measure areas and volumes under functions, but they inherently rely on the assumption that these functions are measurable. This means their level sets (the regions where the function takes values above a certain threshold) must have a well-defined "size" according to the Lebesgue measure.

However, in various advanced mathematical and analytical contexts, functions can be so erratic or "pathological" that they defy Lebesgue measurability. This presented a significant "pain point" for mathematicians: how to extend the powerful tools of integral calculus and related operators (like maximal operators and potential theory) to this broader class of non-measurable functions.

The Choquet integral, introduced by G. Choquet [12], offered a pathway by generalizing the Lebesgue integral to non-additive measures, known as capacities. Initially, even the Choquet integral theory largely focused on functions that were continuous, quasicontinuous, or at least Lebesgue measurable. The crucial shift, and the precise origin of this paper's problem, came from the realization that these measurability assumptions might not always be necessary, especially when considering Choquet integrals with respect to Hausdorff content (which is a type of capacity). D. Denneberg's monograph [16] is highlighted as a key work that extensively studies Choquet integrals without assuming Lebesgue measurability.

This paper builds upon this lineage by explicitly introducing and studying Riesz potentials and maximal operators for Lebesgue non-measurable functions, using the Choquet integral framework with respect to Hausdorff content. Earlier works, such as those by D. R. Adams [2-5] on nonlinear potential theory using Choquet integrals, and more recently by Y.-W. Chen, K. H. Ooi, and D. Spector [11] who introduced the Hausdorff content maximal operator, laid the groundwork. This paper's motivation is to recover or extend these earlier results to the most general setting possible: functions that are not necessarily Lebesgue measurable. The fundamental limitation of previous approaches was their reliance on some form of measurability, which restricted their applicability to a subset of functions, leaving a gap for those that are truly non-measurable. This paper aims to fill that gap, providing a more universal framework for these operators.

Intuitive Domain Terms

Here are some specialized terms from the paper, translated into intuitive analogies for a beginner:

• Choquet Integral: Imagine you're trying to assess the "total impact" of a series of events, but the events aren't perfectly independent; they might overlap or influence each other in complex ways. A standard sum (like a Lebesgue integral) assumes perfect independence. The Choquet integral is like a more flexible, "fuzzy" way of summing up these impacts, accounting for their interactions. It's useful when your "measuring stick" (the underlying measure or capacity) isn't strictly additive.
• Hausdorff Content: Think of this as a "minimal covering cost" or a "roughness index" for a set. Instead of just measuring a set's length, area, or volume, Hausdorff content tells you the smallest total "size" (sum of radii raised to a power $\delta$) of a collection of tiny balls needed to completely cover that set. A very intricate, "fractal-like" set might have a high Hausdorff content for a small $\delta$, even if its standard area is zero, because it's hard to cover efficiently.
• Non-measurable function: Picture a function that is so incredibly erratic and "wiggly" that you can't even define its "average height" or "total area" using conventional methods. It's like trying to weigh a cloud – its boundaries are too ill-defined for a precise measurement with a standard scale. These functions exist in mathematics and pose a challenge for traditional calculus.
• Maximal Operator: This is like a "local peak detector" for a function. For every point in space, it looks at all possible neighborhoods (balls) around that point and finds the highest "average value" of the function within any of those neighborhoods. It essentially tells you how "locally large" a function can get, even if it's small on average globally.
• Riesz Potential: Imagine you have a distribution of "stuff" (like heat or mass) spread out in space. The Riesz potential at a particular point tells you the total "influence" or "potential" exerted by all that "stuff" on that point. The further away the "stuff" is, the less influence it has, and this decay follows a specific mathematical rule. It's like calculating the total gravitational pull at a point from a distributed mass, but in a more generalized way.

Notation Table

| Notation | Description

Problem Definition & Constraints

Core Problem Formulation & The Dilemma

The fundamental problem addressed by this paper is the extension of classical integral operator theories, specifically for Riesz potentials and maximal operators, to a broader class of functions that are not necessarily Lebesgue measurable.

Input/Current State:
Traditional integral theories, such as Riemann and Lebesgue integration, are inherently built upon the assumption that functions are Lebesgue measurable. Similarly, the Choquet integral theory, while more general, has predominantly been applied to functions that exhibit some form of regularity, such as continuity, quasicontinuity, or at least Lebesgue measurability. Consequently, the boundedness properties of key operators like the Hardy-Littlewood maximal operator and Riesz potentials have been established primarily within this measurable context.

Desired Endpoint (Output/Goal State):
The paper aims to precisely define and establish boundedness results for Riesz potentials and maximal operators when applied to functions that are Lebesgue non-measurable. This involves leveraging the Choquet integral framework, particularly with respect to Hausdorff content, which serves as a generalized "measure" or capacity. The goal is to recover or extend existing results from the Lebesgue measurable domain to this more general, non-measurable setting, thereby providing a more comprehensive analytical toolset.

Exact Missing Link or Mathematical Gap:
The precise mathematical gap lies in the absence of a rigorous and widely applicable theory for integral operators (like Riesz potentials and maximal operators) that can handle functions lacking Lebesgue measurability. While the Choquet integral itself can be defined for any non-negative function (Section 3, equation (3.1)), the analytical machinery to prove boundedness inequalities for operators acting on these non-measurable functions, especially when the underlying "measure" is Hausdorff content, was largely undeveloped. The paper seeks to bridge this by formulating these operators and proving their boundedness in appropriate function spaces, such as the newly introduced $NL^p(\Omega, H^\delta_\infty)$ spaces for non-measurable functions.

Painful Trade-off or Dilemma:
The central dilemma that has historically "trapped" researchers is the analytical tractability afforded by Lebesgue measurability versus the desire for a more general theory. Relaxing the measurability requirement introduces significant challenges, as many powerful tools from classical analysis (e.g., standard convergence theorems, Fubini's theorem, properties of $L^p$ spaces) are no longer directly applicable. The trade-off is between the relative ease of working with measurable functions and the need to analyze a wider, more realistic class of functions that may not conform to this strict condition. Previous attempts to generalize often required strong regularity assumptions (like quasicontinuity), which this paper seeks to avoid for maximal operators (Section 3).

Constraints & Failure Modes

The problem of extending integral operator theory to Lebesgue non-measurable functions, particularly with Choquet integrals and Hausdorff content, is fraught with several harsh, realistic walls:

1. Non-Measurability as a Core Constraint: The most significant constraint is the very nature of the functions under consideration: they are not necessarily Lebesgue measurable. This immediately renders many standard analytical techniques invalid. For instance, the definition of the Choquet integral (3.1) is given for any non-negative function, but extending boundedness results for operators like Riesz potentials to this broader class is inherantly difficult, requiring new approaches to function spaces and inequalities.

2. Non-Additivity and Quasi-Subadditivity: Choquet integrals are "nonlinear integrals and used in non-additive measure theory" (Section 3). Unlike Lebesgue integrals, they are generally not additive but rather quasi-subadditive (Lemma 3.1, property (I6)). Furthermore, the Hausdorff content $H^\delta_\infty$ is an outer capacity, satisfying properties (C1)-(C5) but not (C6) (Remark 2.1), meaning it lacks the full additivity properties of a Choquet capacity. This non-additive nature of the underlying "measure" and integral makes proofs of operator boundedness substantially more complex than in classical measure theory, where linearity and additivity are cornerstones. The quasi-norm in $NL^p(\Omega, H^\delta_\infty)$ (Proposition 3.7) reflects this, as its triangle inequality involves a constant $c > 1$, indicating it is not a true norm.

3. Lack of Quasicontinuity Assumption: The authors explicitly state a limitation: "We consider function spaces which are formed by Choquet integrals with respect to Hausdorff content without assuming quasicontinuity. But we are not able to obtain convergence results." (Section 3). Quasicontinuity is a common assumption in the study of Choquet integrals and capacities, facilitating certain regularity and convergence properties. Dropping this assumption, while generalizing the theory, creates a significant analytical hurdle, leading to a recognized "failure mode" where convergence results cannot be readily obtained.

4. Intricate Operator Definitions and Estimates: The definitions of the Hausdorff content maximal operator $M^\delta_\kappa f(x)$ (4.1) and Riesz potential $R^\alpha f(x)$ (5.1) involve complex expressions with suprema over balls and integrals over distance-weighted function values. Proving boundedness for these operators requires highly intricate estimates, often relying on specific properties of Choquet integrals like monotonicity (I4), quasi-subadditivity (I6), and a generalized Hölder's inequality (I7), which differ from their classical counterparts.

5. Parameter Dependencies and Boundary Conditions: The constants in the derived boundedness inequalities (e.g., Theorem 4.3, Theorem 4.7, Theorem 5.2) depend on multiple parameters ($n, \delta, \kappa, p, \alpha$). Managing these dependencies and ensuring the constants remain finite under various parameter ranges (e.g., $p \in (\delta/n, \infty)$, $\kappa \in [0, \delta)$, $\alpha \in (0, \delta)$) adds considerable complexity to the analysys. For instance, Remark 4.4 (2) highlights that the constant $c$ in (4.4) "blows up as $p \to \delta/n$," indicating critical boundary conditions that must be carefully handled.

Why This Approach

The Inevitability of the Choice

The primary motivation for adopting the Choquet integral with respect to Hausdorff content stems from a fundamental limitation of traditional integral theories. The Riemann and Lebesgue integral theories are fundamentaly built upon the assumption that functions are Lebesgue measurable. However, the core problem addressed in this paper involves extending the definitions and boundedness results of Riesz potentials and maximal operators to functions that are not necessarily Lebesgue measurable. The authors explicitly state this realization in the introduction: "The Riemann and Lebesgue integral theories were developed for Lebesgue measurable functions. Ways to study these integral theories when functions are not Lebesgue measurable appear in [24, 26, 46]." They further elaborate that while Choquet integral theory has often been applied to continuous or at least Lebesgue measurable functions, "this seems not to be necessary in many cases when considering properties of Choquet integrals with respect to Hausdorff content." The work of D. Denneberg [16] is highlighted for studying Choquet integrals "extensively without assuming that functions are Lebesgue measurable." This makes the Choquet integral, particularly when paired with Hausdorff content, the only viable solution for performing meaningful analysys on non-measurable functions, as standard Lebesgue integration simply cannot be applied in such a general context.

Comparative Superiority

The Choquet integral with respect to Hausdorff content offers qualitative superiority primarily through its ability to operate on functions and sets that lack Lebesgue measurability. This is a structural advantage over previous gold standards, which are inherently restricted to measurable domains.
* Domain Extension: Unlike Riemann or Lebesgue integrals, the Choquet integral, as defined in Equation (3.1), is "well defined for any Lebesgue non-measurable set in $\mathbb{R}^n$ and for any Lebesgue non-measurable function." This dramatically expands the class of functions amenable to integral analysis, moving beyond the limitations of classical measure theory.
* Non-Additive Framework: The paper notes that the "Choquet integral is a nonlinear integral and used in non-additive measure theory." This allows for a more flexible mathematical framework, which can model phenomena where the traditional additivity of measures might not hold, providing a richer theoretical foundation.
* Generalization and Recovery: The approach allows for the recovery and extension of earlier results. For instance, Theorem 5.5 recovers classical results for Lebesgue measurable functions, while Theorem 5.2 extends these to the broader class of non-measurable functions. This demonstrates that the chosen method is not just an alternative but a more general and encompassing framework.

Alignment with Constraints

The chosen method perfectly aligns with the implicit constraints of the problem, which center on the need to analyze operators for functions that are not necessarily Lebesgue measurable.
* Direct Applicability to Non-Measurable Functions: The most critical constraint is directly met by the Choquet integral's definition. The authors explicitly emphasize that "there is no need to assume that functions are Lebesgue measurable" when using Choquet integrals with respect to Hausdorff content. This is the direct "marriage" between the problem's harsh requirement (non-measurability) and the solution's unqiue property.
* Foundation for New Function Spaces: The paper introduces the quasi-normed space $NL^p(\Omega, H^\delta_\infty)$ specifically for "functions that are not necessarily Lebesgue measurable." This demonstrates that the Choquet integral provides the necessary mathematical machinery to even define and work within function spaces tailored to the problem's non-measurable nature, which is crucial for developing a coherent theory.

Rejection of Alternatives

Within the context of real analysis and measure theory, the primary "alternatives" are traditional integral theories. The paper implicitly rejects these for the problem at hand due to their inherent limitations:
* Riemann and Lebesgue Integrals: These are unsuitable because they are "developed for Lebesgue measurable functions." Since the paper's focus is on "non-measurable functions," these classical methods are fundamentally inapplicable without the very property (measurability) that the problem seeks to circumvent.
* Choquet Integral with Other Capacities/Measures: While the Choquet integral can be defined with respect to various capacities, the specific choice of Hausdorff content is crucial. The authors note that "the Choquet integral theory has been concentrated to the context when functions are continuous or quasicontinuous or at least Lebesgue measurable. However, this seems not to be necessary in many cases when considering properties of Choquet integrals with respect to Hausdorff content." This suggests that other capacities might impose stronger regularity conditions (like continuity or quasicontinuity) on functions, which the Hausdorff content-based Choquet integral avoids, thus providing a more general framework for non-measurable functions.

Mathematical & Logical Mechanism

The Master Equation

The absolut core equation powering the analysis of the Hausdorff content maximal operator in this paper is the definition of the Hausdorff content centred fractional maximal function. This operator extends classical maximal functions to handle non-measurable functions using Choquet integrals and Hausdorff content. It is defined as:

$$ M^\delta_\kappa f(x) := \sup_{r>0} \frac{r^\kappa}{H^\delta_\infty(B(x,r))} \int_{B(x,r)} |f(y)| \, dH^\delta_\infty(y) $$

Term-by-Term Autopsy

Let's dissect this equation piece by piece to understand its components and their roles.

• $M^\delta_\kappa f(x)$: This is the Hausdorff content centred fractional maximal function itself, evaluated at a point $x \in \mathbb{R}^n$.
  • Mathematical Definition: It represents a value associated with the function $f$ at point $x$, capturing the largest weighted average of $f$ in any ball centered at $x$.
  • Physical/Logical Role: This operator quantifies the "local maximal size" or "concentration" of the function $f$ around $x$. It's a generalization of the well-known Hardy-Littlewood maximal operator, adapted for non-measurable functions and fractional orders.
• $\sup_{r>0}$: This denotes the supremum (the least upper bound) over all possible radii $r > 0$.
  • Mathematical Definition: The smallest number that is greater than or equal to all values in a given set.
  • Physical/Logical Role: The "maximal" aspect of the operator comes from this supremum. It ensures that the operator captures the highest possible weighted average of $f$ over any ball centered at $x$, regardless of its size. This is crucial for identifying regions where the function $f$ has significant local "mass." The choice of supremum over, say, an average, is inherent to the definition of a maximal operator, which seeks the "peak" local behavior.
• $r^\kappa$: This is the radius $r$ raised to the power $\kappa$.
  • Mathematical Definition: $r$ is the radius of the ball $B(x,r)$, and $\kappa$ is a parameter such that $\kappa \in [0, \delta)$.
  • Physical/Logical Role: This term introduces the "fractional" nature of the operator. When $\kappa = 0$, it reduces to a standard maximal operator. For $\kappa > 0$, it acts as a scaling factor that modifies the influence of the ball's size on the average. It's akin to the scaling factors seen in fractional integral operators (like Riesz potentials), giving the operator a "long-range" or "smoothing" characteristic depending on the context.
• $H^\delta_\infty(B(x,r))$: This is the $\delta$-dimensional Hausdorff content of the open ball $B(x,r)$.
  • Mathematical Definition: For a set $E \subset \mathbb{R}^n$, $H^\delta_\infty(E) := \inf \{ \sum_{i=1}^\infty r_i^\delta : E \subset \bigcup_{i=1}^\infty B(x_i, r_i) \}$, where the infimum is taken over all countable collections of balls covering $E$. For a ball $B(x,r)$, Proposition 2.2 states $H^\delta_\infty(B(x,r)) = \gamma_\delta r^\delta$ for some constant $\gamma_\delta$.
  • Physical/Logical Role: This term serves as the "normalization factor" or "denominator" for the integral. Instead of using the standard Lebesgue measure (volume), which requires functions to be Lebesgue measurable, the paper uses Hausdorff content. This is a key innovation, allowing the operator to be defined for functions that are not necessarily Lebesgue measurable. The parameter $\delta$ reflects the "dimension" of this content, which can be non-integer.
• $\int_{B(x,r)} |f(y)| \, dH^\delta_\infty(y)$: This is the Choquet integral of the absolute value of the function $f$ over the ball $B(x,r)$ with respect to the $\delta$-dimensional Hausdorff content.
  • Mathematical Definition: For a non-negative function $g: \Omega \to [0, \infty)$, the Choquet integral is defined as $\int_\Omega g(y) \, dH := \int_0^\infty H(\{y \in \Omega : g(y) > t\}) \, dt$. Here, $g(y) = |f(y)|$, $\Omega = B(x,r)$, and $H = H^\delta_\infty$.
  • Physical/Logical Role: This is the "accumulation" or "summation" part of the operator. It measures the "total amount" of the function $|f|$ within the ball $B(x,r)$. The use of the Choquet integral is fundamental because it allows integration with respect to non-additive set functions like Hausdorff content, which is essential for handling non-measurable functions. The absolute value $|f(y)|$ is used because the Choquet integral is typically defined for non-negative functions. The integral is used to compute an average, as it is the standard tool for aggregating function values over a domain.
• $f(y)$: The input function.
  • Mathematical Definition: A function $f: \mathbb{R}^n \to [-\infty, \infty]$.
  • Physical/Logical Role: This is the function whose local behavior (its maximal average) is being analyzed. The paper's focus is on functions that are not necessarily Lebesgue measurable.
• $y$: The integration variable.
  • Mathematical Definition: A point in $\mathbb{R}^n$.
  • Physical/Logical Role: Represents the points within the ball $B(x,r)$ over which the function $f$ is being integrated.
• $x$: The center point.
  • Mathematical Definition: A point in $\mathbb{R}^n$.
  • Physical/Logical Role: This is the specific location in space where the maximal operator is being evaluated. The balls $B(x,r)$ are "centred" at this point.
• $B(x,r)$: An open ball centered at $x$ with radius $r$.
  • Mathematical Definition: The set of all points $y \in \mathbb{R}^n$ such that the Euclidean distance $|x-y|$ is less than $r$.
  • Physical/Logical Role: This defines the local neighborhood around $x$ over which the function $f$ is being averaged.

Step-by-Step Flow

Imagine a mechanical assembly line processing the function $f$ to produce the maximal operator $M^\delta_\kappa f(x)$ at a given point $x$.

1. Point Selection: We start by fixing a specific point $x$ in $\mathbb{R}^n$ where we want to compute the value of $M^\delta_\kappa f(x)$. This $x$ is our observation point.
2. Radius Iteration (The Scanner): A conceptual "scanner" at point $x$ begins to expand its reach, considering an infinite sequence of possible radii $r > 0$. For each radius $r$, a new "local processing unit" is activated.
3. Ball Definition: For each chosen radius $r$, an open ball $B(x,r)$ is formed, centered at $x$ with that radius. This ball defines the current local neighborhood.
4. Absolute Value Transformation: Inside this ball $B(x,r)$, the function $f(y)$ is processed. For every point $y$ within the ball, its value $f(y)$ is passed through an "absolute value gate," yielding $|f(y)|$. This ensures all values are non-negative, a requirement for the next step.
5. Choquet Integration (The Accumulator): The non-negative values $|f(y)|$ within $B(x,r)$ are then fed into a specialized "Choquet integral accumulator." This accumulator works by:
   • For each possible positive threshold $t$, it identifies all points $y$ in $B(x,r)$ where $|f(y)|$ exceeds $t$. This forms a set $E_t$.
   • It then measures the "size" of this set $E_t$ using the $\delta$-dimensional Hausdorff content, $H^\delta_\infty(E_t)$.
   • Finally, it integrates these content values $H^\delta_\infty(E_t)$ over all possible thresholds $t$ from $0$ to $\infty$. This yields a single numerical value: $\int_{B(x,r)} |f(y)| \, dH^\delta_\infty(y)$.
6. Content Normalization: Simultaneously, the $\delta$-dimensional Hausdorff content of the entire ball $B(x,r)$, denoted $H^\delta_\infty(B(x,r))$, is calculated. This value is then used to normalize the Choquet integral from the previous step.
7. Fractional Scaling: The normalized value is then multiplied by $r^\kappa$, where $\kappa$ is a predetermined fractional parameter. This scales the result based on the ball's radius and the fractional order.
8. Value Storage: The resulting scaled and normalized Choquet average for the current radius $r$ is stored.
9. Maximization (The Selector): After all possible radii $r$ have been processed (steps 2-8), a "selector" mechanism reviews all the stored values. It picks out the largest value among them. This largest value is the final output of the assembly line for the point $x$, which is $M^\delta_\kappa f(x)$. This process is repeated for every point $x$ in $\mathbb{R}^n$ to define the entire maximal function.

Optimization Dynamics

It's important to clarify that the Hausdorff content maximal operator, $M^\delta_\kappa f(x)$, as defined and studied in this paper, is a theoretical mathematical construct within harmonic analysis and potential theory. It is not an algorithm or a model that "learns," "updates," or "converges" in the sense of iterative optimization, machine learning, or numerical methods.

There are no gradients computed, no loss landscapes shaped, and no iterative state updates over time. The operator is a direct definition that, for any given function $f$ and point $x$, yields a specific value. The paper's focus is on proving properties of this operator, such as its boundedness on certain function spaces (e.g., Theorem 4.3), its lower semicontinuity (Proposition 4.2), and its relationships with other operators. These are analytical results about the operator's behavior, not about an optimization process. Therefore, the concept of "optimization dynamics" as typically understood in computational or statistical contexts does not apply here.

Results, Limitations & Conclusion

Experimental Design & Baselines

This paper is a work of pure mathematics, focusing on theoretical proofs rather than empirical experimental validation. Therefore, there isn't an "experimental design" in the traditional sense of running simulations or collecting data. Instead, the authors architect their proofs by rigorously defining new mathematical objects—the Hausdorff content maximal operator and Riesz potential for non-measurable functions—and then demonstrating their properties, primarily boundedness, through logical deduction and the application of established theorems from measure theory, potential theory, and functional analysis.

The "baselines" or "victims" in this context are the classical results and operators defined for Lebesgue measurable functions. The paper's core contribution is to extend these classical concepts to the realm of non-measurable functions by leveraging Choquet integrals with respect to Hausdorff content. For instance, the classical Hardy-Littlewood maximal operator and Riesz potential are well-understood for Lebesgue measurable functions. The authors "defeat" the limitations of these classical frameworks by showing that their newly defined operators for non-measurable functions either recover these classical results under specific conditions (e.g., when functions are Lebesgue measurable) or extend them to a much broader class of functions. This extension is the definitive, undeniable evidence that their core mechanism—the Choquet integral with Hausdorff content—actually works in reality, in the mathematical sense of providing a consistent and useful generalization.

What the Evidence Proves

The central mechanism proven effective in this paper is the successful definition and analysis of maximal operators and Riesz potentials for functions that are not necessarily Lebesgue measurable, using the Choquet integral with respect to Hausdorff content. The evidence for this success is presented through a series of theorems and propositions establishing crucial properties, most notably boundedness results.

A key piece of evidence is Theorem 4.3, which proves the boundedness of the Hausdorff content maximal operator $M^\delta f(x)$ for non-Lebesgue measurable functions. Specifically, it shows that for $p \in (\delta/n, \infty)$, there exists a constant $c$ such that
$$ \int_{R^n} (M^\delta f(x))^p dH^\delta_\infty \le c \int_{R^n} |f(x)|^p dH^\delta_\infty $$
for all functions $f: R^n \to [-\infty, \infty]$. This is a direct extension of classical maximal function theory to a more general setting. A crucial intermediate result, Proposition 4.2, provides foundational support by demonstrating that, remarkably, the maximal function $M^\delta f$ itself is Lebesgue measurable, even when the original function $f$ is not. This ensures that the integral on the left-hand side of the inequality is well-defined in the Lebesgue sense.

Further evidence is provided by Theorem 4.7, which extends this boundedness to the fractional maximal operator $M^\delta_\kappa f(x)$ for non-Lebesgue measurable functions under specific parameter conditions. Similarly, Proposition 4.11 establishes boundedness for the Hausdorff content sharp maximal function $M^\# f(x)$, again for non-Lebesgue measurable functions.

For the Riesz potential, Theorem 5.2 is a cornerstone, proving its boundedness for non-Lebesgue measurable functions. It states that for $p \in (\delta/n, \delta/\alpha)$,
$$ \left(\int_{R^n} (R^\alpha f(x))^{\frac{\delta p}{\delta - \alpha p}} dH^\delta_\infty\right)^{\frac{\delta - \alpha p}{\delta p}} \le c \left(\int_{R^n} |f(x)|^p dH^\delta_\infty\right)^{\frac{1}{p}} $$
where $c$ is a constant. This inequality demonstrates that the Riesz potential, when defined via Choquet integrals and Hausdorff content, maintains a controlled behavior analogous to its classical counterpart. The paper also shows that for Lebesgue measurable functions, Theorem 5.5 and Theorem 5.6 recover and extend classical results, such as those found in [47, Theorem 2.8.4], by incorporating the Hausdorff content framework and additional parameters.

The definitive evidence is the consistent mathematical proof that these newly defined operators, which are designed to handle functions beyond the scope of traditional Lebesgue integration, exhibit fundamental properties like boundedness. This extends the applicability of these analytical tools to a wider array of mathematical problems where non-measurable functions naturally occure.

Limitations & Future Directions

While this paper makes significant theoretical strides, it also highlights several limitations and opens avenues for future research.

One explicit limitation is the absence of convergence results for the introduced function spaces $NL^p(\Omega, H^\delta_\infty)$ (Page 7). While these spaces are defined, the lack of convergence properties can hinder their utility in dynamic or approximation-based analyses. Relatedly, the quasi-norm defined for these spaces is not necessarily a true norm, as the quasi-subadditivity inequalities (I6) and (I7) might not hold with a constant of one. Understanding the implications of this, or finding conditions under which it becomes a true norm, is an important future direction.

Some results, such as Theorem 4.9, are restricted to H-quasicontinuous functions, which is a stronger condition than merely non-measurable. Investigating whether these results can be extended to broader classes of non-measurable functions would be valuable. Additionally, Remark 4.5 points out that the proof technique for Theorem 4.3 (maximal operator) does not directly apply to the fractional maximal operator $M^\delta_\kappa f$ when $\delta < n$ and $\kappa > 0$, suggesting a gap in the current framework for this specific case. The constant $c$ in Theorem 4.3 is noted to blow up as $p \to \delta/n$ (Remark 4.4(2)), indicating a potential singularity or boundary condition that warrants further investigation for stability and behavior near this limit.

Looking forward, several discussion topics emerge:

• Establishing Convergence Theory: A critical next step is to develop a robust convergence theory for the $NL^p(\Omega, H^\delta_\infty)$ spaces. This would significantly enhance their analytical power, enabling the study of sequences of non-measurable functions and their limits, which is fundamental in many areas of analysis and applications.
• Applications in Diverse Fields: Given the ability to handle non-measurable functions, how can these new operators be applied to problems in fields like signal processing, image analysis, or even theoretical physics, where functions with highly irregular or fractal-like behavior are common? Could they offer new insights into phenomena that are difficult to model with classical Lebesgue theory?
• Refining Inequalities and Constants: The paper notes that some inequalities might be valid for larger classes of functions or that the dimension of Hausdorff content on the left-hand side could be smaller than on the right-hand side (Remarks 5.3(1), 5.4). Future work could focus on tightening these bounds, finding sharper constants, or relaxing the parameter dependance.
• Connections to Other Capacities and Measures: The analysis primarily uses Hausdorff content. Exploring how these definitions and boundedness results translate or generalize to other types of capacities or non-additive measures (e.g., fuzzy measures, belief functions) could lead to a more unified theory of integrals for irregular functions.
• Development of Numerical Methods: While theoretical, the concepts could inspire novel numerical methods for approximating integrals or analyzing operators involving highly irregular functions. This could have practical implications for computational mathematics and scientific computing.
• Exploring Limit Cases and Boundary Behaviors: A deeper dive into the implications and applications of the limit cases, such as inequality (5.6) for compactly supported functions, could reveal interesting properties or connections to other mathematical theories. Understanding the behavior of constants near critical parameter values (e.g., $p \to \delta/n$) is also crucial for a complete theoretical understanding.

Connections to Other Fields

Mathematical Skeleton

The pure mathematical core of this work involves the development of integral operators (maximal and potential types) within a non-additive integration framework, where the underlying 'measure' is a capacity, and establishing their boundedness properties in associated function spaces. This framework allows for the analysis of functions where traditional Lebesgue measurability assumptions are not met, with key operations occuring via Choquet integrals.

Adjacent Research Areas

Non-additive Measure Theory and Decision Theory

This paper draws heavily from and contributes to the field of non-additive measure theory, particularly concerning the Choquet integral. The Choquet integral, introduced by G. Choquet [12, 1953, Ann. Inst. Fourier (Grenoble)], provides a way to integrate functions with respect to capacities or fuzzy measures, which are set functions that are not necessarily additive. The paper's definition of the Riesz potential and maximal operator relies entirely on the Choquet integral (3.1), extending its application to non-measurable functions. The properties of Choquet integrals, such as quasi-sublinearity (3.3) and Hölder's inequality (I7), are fundemental to the proofs of boundedness results for the newly defined operators.

Potential Theory and Harmonic Analysis

The concepts of maximal operators and Riesz potentials are cornerstones of classical potential theory and harmonic analysis. This work generalizes these classical operators, which are typically defined using Lebesgue integrals, to a setting involving Choquet integrals and Hausdorff content. For instance, the Hausdorff content Riesz potential (5.1) and the Hausdorff content maximal operator (4.1) are direct analagous extensions of their classical counterparts. The boundedness results presented (e.g., Theorem 4.3 for maximal operators and Theorem 5.2 for Riesz potentials) extend the well-known $L^p$ boundedness theorems from classical analysis to these more general, non-measurable function spaces. A representative work connecting Choquet integrals to potential theory is D.R. Adams [3, 1998, Publ. Mat.].

Geometric Measure Theory and Fractal Geometry

The entire analytical framework of this paper is built upon the $\delta$-dimensional Hausdorff content $H^\delta_\infty$ (2.1), a key concept in geometric measure theory and fractal geometry. Hausdorff content and its related Hausdorff measure are essential tools for quantifying the "size" of sets, especially those with non-integer dimensions or complex geometric structures. By replacing the Lebesgue measure with Hausdorff content as the underlying capacity for Choquet integrals, the paper extends the study of maximal operators and Riesz potentials to functions defined on sets that might exhibit fractal properties or where the standard Euclidean dimension is not the most appropriate measure. This approach allows for a more nuanced analysis of functions in irregular settings. A foundational text in this area is H. Federer [18, 1969, Springer, New York].