2025-11-10T02:52:56.917839

Well-posedness of aggregation-diffusion systems with irregular kernels

Carrillo, Salmaniw, Skrzeczkowski
We consider aggregation-diffusion equations with merely bounded nonlocal interaction potential $K$. We are interested in establishing their well-posedness theory when the nonlocal interaction potential $K$ is neither differentiable nor positive (semi-)definite, thus preventing application of classical arguments. We prove the existence of weak solutions in two cases: if the mass of the initial data is sufficiently small, or if the interaction potential is symmetric and of bounded variation without any smallness assumption. The latter allows one to exploit the dissipation of the free energy in an optimal way, which is an entirely new approach. Remarkably, in both cases, under the additional condition that $\nabla K\ast K$ is in $L^2$, we can prove that the solution is smooth and unique. When $K$ is a characteristic function of a ball, we construct the classical unique solution. Under additional structural conditions we extend these results to the $n$-species system.
academic

Well-posedness of aggregation-diffusion systems with irregular kernels

Basic Information

  • Paper ID: 2406.09227
  • Title: Well-posedness of aggregation-diffusion systems with irregular kernels
  • Authors: José A. Carrillo, Yurij Salmaniw, Jakub Skrzeczkowski
  • Classification: math.AP (Analysis of PDEs)
  • Publication Date: June 2024 (arXiv preprint, revised October 2024)
  • Paper Link: https://arxiv.org/abs/2406.09227

Abstract

This paper investigates the well-posedness theory for aggregation-diffusion equations with bounded nonlocal interaction potentials KK. When the nonlocal interaction potential KK is neither differentiable nor positive (semi)definite, classical methods no longer apply. The authors prove the existence of weak solutions in two scenarios: (1) when the initial data mass is sufficiently small; (2) when the interaction potential is symmetric with bounded variation, without requiring smallness assumptions. The latter approach optimally exploits the dissipation of free energy through a novel methodology. Under the additional condition ∇K∗K∈L2\nabla K * K \in L^2, regularity and uniqueness of solutions can be established. Classical unique solutions are constructed when KK is the characteristic function of a ball. These results are generalized to nn-population systems under additional structural conditions.

Research Background and Motivation

Importance of the Problem

Aggregation-diffusion equations serve as mean-field approximations for interacting agents or particles across numerous scientific disciplines, including:

  • Fluid dynamics, environmental science, and chemical engineering
  • Phase separation in materials science
  • Cell-cell adhesion in biology
  • Collective dynamics, biological aggregation, and ecology

Limitations of Existing Approaches

Existing well-posedness theories primarily address:

  1. Smooth kernels: requiring K∈W2,∞K \in W^{2,\infty} or twice continuously differentiable KK
  2. Positive definite kernels: exploiting entropy dissipation, requiring positive definiteness and detailed balance conditions
  3. Singular kernels: such as Newtonian potentials in Keller-Segel systems

Research Motivation

Many kernels arising in practical applications (such as top-hat kernels) are neither smooth nor positive definite, rendering existing theory inapplicable. The core motivation of this work is:

  • Resolving open problems posed in 73
  • Verifying conjectures from 70
  • Establishing robust well-posedness theory for low-regularity kernels

Core Contributions

  1. Establishes well-posedness theory for irregular kernels: First treatment of bounded kernels that are neither differentiable nor positive definite
  2. Proposes two novel existence proof strategies:
    • Entropy method for small mass cases
    • Free energy dissipation method based on gradient flow structure for arbitrary mass
  3. Proves uniqueness of strong solutions: under the condition ∇K∗K∈L2\nabla K * K \in L^2
  4. Constructs classical solutions: for the characteristic function of a ball case
  5. Extends to multi-population systems: nn-population case under detailed balance conditions

Detailed Methodology

Problem Formulation

Investigation of well-posedness for the following aggregation-diffusion system:

Scalar equation:

\frac{\partial u}{\partial t} = \nabla \cdot (D\nabla u + u\nabla(K * u)) \\ u(0,x) = u_0(x) \end{cases}$$ **$n$-population system**: $$\begin{cases} \frac{\partial u_i}{\partial t} = \nabla \cdot \left(D_i\nabla u_i + u_i\sum_{j=1}^n \nabla(K_{ij} * u_j)\right) \\ u_i(0,x) = u_{i0}(x) \end{cases}$$ where $K * v(t,x) := \int_{\mathbb{R}^d} K(x-y)v(t,y)dy$ denotes spatial convolution. ### Core Assumptions **(H1)** $K_{ij} \in L^1(\mathbb{R}^d) \cap L^{\infty}(\mathbb{R}^d)$ **(H2)** $K_{ij} \in BV(\mathbb{R}^d)$, i.e., $\|\nabla K_{ij}\|_{TV} < \infty$ **(H3)** $K_{ij}$ is symmetric about the origin **(H4)** $\nabla(\tilde{K}_{ij} * K_{ij}) \in L^2(\mathbb{R}^d)$ **(H5)** Detailed balance condition: there exist constants $\pi_i > 0$ such that $\pi_i K_{ij} = \pi_j K_{ji}$ **(H6)** $K_{ij}$ has compact support ### Technical Innovations #### 1. Entropy Estimation Method for Small Mass Cases For the entropy functional $H[u] = \int u \log u dx$, the following is proven: $$\frac{d}{dt}H[u(t)] + 4c_{1,1}\|\nabla\sqrt{u(t)}\|_{L^2}^2 \leq 0$$ where $c_{1,1} = D - m\|K\|_{L^{\infty}} > 0$. Key techniques: - Controlling nonlocal terms via Hölder inequality and Young convolution inequality - Controlling the negative part of entropy through second moment bounds - Obtaining global estimates via Grönwall's lemma #### 2. Gradient Flow Method for Arbitrary Mass Cases Reformulating the equation as a gradient flow: $$u_t = \nabla \cdot (\sqrt{u}f), \quad f := \sqrt{u}\nabla(D\log u + K*u)$$ Utilizing the free energy functional: $$F[u] = D\int u\log u dx + \frac{1}{2}\int u(K*u)dx$$ Proving the dissipation property: $$\frac{d}{dt}F[u(t)] + \|f(t)\|_{L^2}^2 = 0$$ #### 3. Higher-Order Regularity Estimates Under condition (H4), improved estimates are obtained via test function $\tilde{K}*K*\Delta u$: $$\frac{1}{2}\frac{d}{dt}\int|\nabla(K*u)|^2 dx + D\int|\Delta(K*u)|^2 dx \leq C(t)$$ ## Main Results ### Theorem 2.1 (Small Mass Weak Solution Existence) Assume (H1) holds and $m\|K\|_{L^{\infty}} < D$. Then there exists a global weak solution. ### Theorem 2.2 (Arbitrary Mass Weak Solution Existence) Assume (H1)-(H3) hold. Then for any initial mass, there exists a global weak solution. ### Theorem 2.3 (Strong Solution Uniqueness) Under the conditions of Theorem 2.1 or 2.2, adding (H4) and appropriate initial data regularity, the weak solution is a unique global strong solution. ### Theorem 2.4 (Classical Solution Existence) Under appropriate initial data regularity and compact support condition (H6), the strong solution is a unique global classical solution that is strictly positive for all $t > 0$. ## Numerical Experiments ### One-Dimensional Top-Hat Kernel Simulations Considering the kernel function: $$K(x) = \begin{cases} -\frac{\alpha}{2R}, & -R \leq x \leq R \\ 0, & \text{otherwise} \end{cases}$$ Parameter settings: $D = 0.25$, $R = 1.0$, $\ell = 4.0$ #### Main Findings: 1. **Weak attraction** ($\alpha = 2$): Diffusion dominates, solution tends toward constant state 2. **Strong attraction** ($\alpha = 30$): Attraction dominates, mass concentrates in a single region 3. **Moderate attraction** ($\alpha = 20$): Bimodal pattern formation 4. **Strong repulsion** ($\alpha = -20$): Pattern state formation with slow decay over long times ### Two-Population System Investigation of two-population systems satisfying detailed balance conditions reveals: - Population segregation phenomena - Complex rearrangement dynamics - Formation of final steady states ## Related Work ### Classical Results - **Smooth kernels**: Classical well-posedness theory for $K \in W^{2,\infty}$ [28,31] - **Positive definite kernels**: Well-posedness results via entropy methods [52] - **Singular kernels**: Keller-Segel systems [7,9,11-14,16,19,21,33,34,37,43,47,53] ### Position of This Work This paper fills the gap between smooth and singular kernels, addressing bounded variation but non-smooth kernels, which remain unexplored in existing literature. ## Conclusions and Discussion ### Main Conclusions 1. Establishes complete well-posedness theory for aggregation-diffusion equations with irregular kernels 2. Provides two distinct existence proof strategies applicable to different parameter ranges 3. Proves solution uniqueness and regularity under reasonable conditions 4. Successfully extends to multi-population cases ### Limitations 1. **Nonlinear diffusion**: Methods cannot directly extend to porous medium cases with $m \neq 1$ 2. **Detailed balance condition**: Multi-population cases require strong structural assumptions 3. **Dimensional restrictions**: Some results (e.g., Appendix A) apply only to low-dimensional cases ### Future Directions 1. Investigate more general nonlinear diffusion cases 2. Relax restrictions on detailed balance conditions 3. Explore long-time asymptotic behavior and steady-state existence 4. Study well-posedness in higher dimensions ## In-Depth Evaluation ### Strengths 1. **Theoretical innovation**: First systematic treatment of well-posedness for irregular kernels 2. **Methodological diversity**: Provides two distinct proof strategies, revealing problem richness 3. **Technical depth**: Skillfully combines entropy methods, gradient flow theory, and PDE regularity theory 4. **Completeness**: Comprehensive theoretical framework from weak to classical solutions 5. **Practicality**: Addresses important problems in real applications ### Weaknesses 1. **Technical conditions**: Geometric meaning of certain assumptions (e.g., (H4)) lacks intuitive clarity 2. **Generalization limitations**: Extension to nonlinear cases faces fundamental difficulties 3. **Numerical verification**: Numerical experiments primarily concentrate on one-dimensional cases ### Impact 1. **Theoretical contribution**: Provides new tools for PDE theory in handling low-regularity problems 2. **Application value**: Directly applicable to models in biomathematics, materials science, and related fields 3. **Methodological significance**: Demonstrates how to combine diverse mathematical tools for complex problems ### Applicable Scenarios - Non-smooth perception functions in biological aggregation models - Phase separation problems in materials science - Local interaction models in collective dynamics - Spatial memory models in ecology ## References [28] J. A. Carrillo, R. S. Gvalani, G. A. Pavliotis, and A. Schlichting. Long-time behaviour and phase transitions for the McKean-Vlasov equation on the torus. [52] A. Jüngel, S. Portisch, and A. Zurek. Nonlocal cross-diffusion systems for multi-species populations and networks. [70] J. R. Potts and M. A. Lewis. Spatial memory and taxis-driven pattern formation in model ecosystems. [73] H. Wang and Y. Salmaniw. Open problems in PDE models for knowledge-based animal movement via nonlocal perception and cognitive mapping.