Existence of martingale solutions for a stochastic weighted mean curvature flow of graphs

Basic Information

• Paper ID: 2510.09383
• Title: Existence of martingale solutions for a stochastic weighted mean curvature flow of graphs
• Authors: Qi Yan, Xiang-Dong Li
• Classification: math.AP (Mathematical Analysis), math.PR (Probability Theory)
• Publication Date: October 14, 2025
• Paper Link: https://arxiv.org/abs/2510.09383

Abstract

This paper investigates the stochastic weighted mean curvature flow of graphs with additional forces on periodic domains of arbitrary dimension. Based on the compact embedding method for variational SPDEs, the existence of martingale solutions is established. Furthermore, the small perturbation limit of the stochastic weighted mean curvature flow is derived.

Research Background and Motivation

Problem Background

1. Importance of Geometric Flows: Mean curvature flow (MCF) is an important geometric flow with significant theoretical and applied implications, describing physical phenomena such as crystal growth and two-phase interface evolution.
2. Weighted Mean Curvature Flow: This paper studies the f-mean curvature flow, which is the gradient flow of weighted volume with respect to the weighted measure dμ = e^(-f)dx, where f is a smooth function on ℝ^(n+1).
3. Necessity of Stochastic Perturbations: Kawasaki and Ohta first proposed mean curvature flow with random forces from time-dependent Ginzburg-Landau models to describe the effects of thermal noise.
4. Limitations of Existing Research:
   • Existing studies focus primarily on specific dimensions or special geometric settings
   • Lack of systematic investigation of weighted cases on general periodic domains
   • Need for more precise mathematical frameworks to handle stochastic terms

Research Motivation

This paper aims to establish the existence theory of martingale solutions for stochastic f-mean curvature flow on tori of arbitrary dimension and investigate its small perturbation limit behavior, providing a rigorous mathematical foundation for understanding stochastic geometric flows.

Core Contributions

1. Existence Theorem: Under appropriate pinching conditions, the existence of martingale solutions for stochastic f-mean curvature flow on tori of arbitrary dimension is established.
2. Unified Theoretical Framework: The compact embedding method for variational SPDEs is employed, providing a systematic approach for handling geometrically complex stochastic partial differential equations.
3. Small Perturbation Limit: The convergence of the stochastic f-mean curvature flow to the deterministic f-mean curvature flow as noise intensity λ→0 is established.
4. Energy Estimates: Uniform a priori energy estimates with respect to ε≥0 are obtained, which are crucial for proving solution compactness.

Methodology Details

Problem Formulation

Investigation of the existence of martingale solutions for the following stochastic f-mean curvature flow:

du = (Q(∇u)div_f v(∇u) + ξu)dt + Q(∇u) ◦ dW

where:

• u: T^n → ℝ is the height function
• Q(p) = √(1 + |p|²) is the area element
• v(p) = p/√(1 + |p|²) is the horizontal projection of the inner normal vector
• div_f V = div V - ∇f · V is the divergence with respect to the weighted measure

Model Architecture

1. Viscous Approximation

Introduction of viscous terms to construct approximating equations:

du = (ε∆_f u + Q(∇u)div_f v(∇u) + ξu)dt + Q(∇u) ◦ dW

2. Variational SPDE Framework

• Space Setting: V = {∇u | u ∈ H²(T^n, dμ)}, H = {∇u | u ∈ H¹(T^n, dμ)}
• Operator Definition:
  • A_ε: V → V' as the principal operator
  • B: V → H as the noise operator

3. Compact Embedding Method

Utilization of the following compact embeddings:

• C^{0,λ}(0,T; L²) ∩ C(0,T; H¹) ↪ C(0,T; (H¹,w))
• L²(0,T; H²) ∩ C^{0,λ}(0,T; L²) ↪ L²(0,T; H¹)

Technical Innovations

1. Gradient Method: Direct application of variational SPDE theory to ∇u rather than u, avoiding technical difficulties in traditional approaches.
2. Energy Estimation Techniques: Uniform energy bounds with respect to ε are obtained through selection of special test functions h(Q(∇u)).
3. Maximum Principle: L^∞ boundedness of the gradient is established, which is crucial for solution regularity.
4. Geometric Application of Itô Formula: Stochastic differential formulas applicable to geometric quantities are developed.

Experimental Setup

Theoretical Verification

This paper is primarily theoretical work, with results verified through:

1. Condition Verification: Verification of coercivity, growth, and continuity conditions required by variational SPDE theory
2. Energy Estimates: Detailed calculations verifying energy inequalities
3. Limit Processes: Verification of the validity of limit processes as ε→0 and λ→0

Key Assumptions

• Pinching Condition: D²f ≥ δξId > 0
• Initial Value Condition: Support on {z ∈ H¹: ||∇z||_{L^∞} ≤ L}
• Noise Intensity: λ² < 2 (small perturbation case)

Experimental Results

Main Theoretical Results

Theorem 3.2 (Existence)

Under the pinching condition D²f ≥ δξId > 0, there exists a martingale solution to the stochastic f-mean curvature flow satisfying:

• D²u ∈ L²(Ω; L²(0,∞; L²(T^n, dμ)))
• ||∇u||_{L^∞(0,∞;L^∞)} ≤ L P-a.s.

Proposition 5.1 (Energy Estimates)

Under appropriate conditions, the energy ∫_{T^n} |∇u|²dμ is a supermartingale with precise energy inequalities.

Theorem 7.5 (Small Perturbation Limit)

As λ→0, the stochastic solution u^λ converges almost surely to the solution of the deterministic f-mean curvature flow.

Technical Achievements

1. Uniform Estimates: A priori estimates uniform with respect to ε≥0 and λ²<2 are obtained
2. Regularity: The gradient of the solution is shown to maintain L^∞ boundedness
3. Long-time Behavior: Global existence of solutions is established

Related Work

Main Research Directions

1. Deterministic MCF: Classical work by Huisken, Ecker and others establishing deterministic mean curvature flow theory
2. Stochastic MCF:
   • Yip (1998): First study of stochastic MCF existence
   • Es-Sarhir & von Renesse (2012): Curve shortening flow
   • Hofmanová, Röger & von Renesse (2017): Weak solutions for two-dimensional graphs
   • Dabrock, Hofmanová & Röger (2021): Martingale solutions for arbitrary dimensions

Innovations of This Paper

Compared to existing work, this paper:

1. First addresses stochastic MCF in the weighted case
2. Provides more general dimensions and geometric settings
3. Establishes small perturbation limit theory
4. Develops more refined energy estimation techniques

Conclusions and Discussion

Main Conclusions

1. Under appropriate geometric conditions, stochastic weighted mean curvature flow admits global martingale solutions
2. Solutions preserve the regularity of initial values (gradient boundedness)
3. Small noise limits converge to deterministic flow

Limitations

1. Geometric Restrictions: Only considers the case of graphs, not general hypersurfaces
2. Condition Requirements: The pinching condition D²f ≥ δξId > 0 is relatively strong
3. Uniqueness: Solution uniqueness is not established
4. Long-time Behavior: Asymptotic behavior of solutions is not deeply investigated

Future Directions

1. Investigation of solution uniqueness and path continuous dependence
2. Extension to more general geometric settings
3. Study of long-time asymptotic behavior of solutions
4. Development of numerical methods to verify theoretical results

In-depth Evaluation

Strengths

1. Theoretical Rigor: Complete proofs with precise technical details
2. Methodological Innovation: The variational SPDE method on gradient spaces is original
3. Complete Results: Comprehensive theoretical framework from existence to limit behavior
4. Technical Depth: Energy estimation and compactness arguments are highly technical

Weaknesses

1. Limited Applications: Pure theoretical work lacking numerical verification or application examples
2. Strict Conditions: Pinching conditions may be too strong for practical applications
3. Writing Density: Dense technical content with room for improved readability
4. Geometric Limitations: Restricted to the case of graphs

Impact

1. Academic Value: Important supplement to stochastic geometric flow theory
2. Methodological Contribution: Compact embedding method generalizable to other stochastic PDEs
3. Theoretical Completeness: Fills theoretical gaps in the weighted case
4. Foundation for Future Research: Provides solid foundation for further investigation

Applicable Scenarios

1. Interface evolution modeling in materials science
2. Stochastic diffusion processes in image processing
3. Stochastic interface models in financial mathematics
4. Membrane dynamics research in biology

References

The paper cites 33 important references, primarily including:

• Classical literature on geometric flows (Huisken, Ecker, etc.)
• Stochastic PDE theory (Da Prato & Zabczyk, Prévôt & Röckner, etc.)
• Pioneering work on stochastic MCF (Yip, Hofmanová, etc.)
• Recent developments in variational SPDEs (Dabrock, etc.)

Overall Assessment: This is a high-quality theoretical mathematics paper making important contributions to the field of stochastic geometric flows. Although highly technical, it provides a rigorous mathematical foundation for understanding stochastic curvature flows and possesses significant academic value.