The ℳ \mathcal{M} -Operator and Uniqueness of Nonlinear Kinetic Equations

Ricardo Alonso, Maria Pia Gualdani, Weiran Sun · 2025

Read The ℳ \mathcal{M} -Operator and Uniqueness of Nonlinear Kinetic Equations by Ricardo Alonso, Maria Pia Gualdani, Weiran Sun (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2506-20775.txt.

Introduction

Ricardo Alonso, Maria Pia Gualdani, and Weiran Sun introduce a tool that settles whether kinetic equations have just one solution, in a 2025 paper. Such equations track gases and plasmas through particle collisions, and Landau-type forms are among the hardest. Older proofs of uniqueness demand that the solution be smooth, with bounded slopes, which real solutions may lack.

Their tool, the M-operator, acts like a fractional slope in both space and speed, and it plays well with the flow term. With it, plain continuous or bounded solutions can be shown unique. The method thus frees uniqueness proofs from smoothness demands and widens the class of equations they cover.

Catalog record

CREATORRicardo Alonso, Maria Pia Gualdani, Weiran Sun
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE107 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2506-20775

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

The ℳ \mathcal{M} -Operator and Uniqueness of Nonlinear Kinetic Equations Ricardo Alonso, Maria Pia Gualdani, Weiran Sun Abstract . We introduce an ℳ \mathcal{M} -operator approach to establish the uniqueness of continuous or bounded solutions for a broad class of Landau-type nonlinear kinetic equations. The specific ℳ \mathcal{M} -operator, originally developed in [ 3 ] , acts as a negative fractional derivative in both spatial and velocity variables and interacts in a controllable manner with the kinetic transport operator. The novelty of this method is that it bypasses the need for bounds on the derivatives of the solution - an assumption typically required in uniqueness arguments for non-cutoff equations. As a result, the method enables working with solutions with low regularity. 1. Introduction The uniqueness of solutions with low regularity or, equivalently, with rough initial data is a central and challenging question in the analysis of nonlinear kinetic equations. There are several unsolved problems in the context of uniqueness; one of the most relevant is the uniqueness of renormalized solution for the Boltzmann equation [ 6 ] and uniqueness of entropic solutions (also called H solutions) [ 24 ] for homogeneous Boltzmann and Landau equations. Homogeneous equations are, in general, simpler to treat and give important insights about the inhomogeneous counterpart. In this setting, the most general results of uniqueness for homogeneous non-cutoff Boltzmann and Landau problems can be found in [ 9 ] and [ 11 ] , respectively, with an analysis based on the Wasserstein distance. In addition, the recent work [ 14 ] settles the question of uniqueness of solution to the homogeneous Landau-Coulomb equation for initial data in polynomially weighted negative Sobolev spaces. These techniques, however, are not easily adaptable to the inhomogeneous case. In the context of the inhomogeneous cutoff Boltzmann equation close to equilibrium, one of the most general uniqueness results can be found in [ 13 ] , in particular, one related to the uniqueness of bounded solutions. A counterpart in the context of non-cutoff Boltzmann or Landau equations is an open problem in the area. This is a key motivating factor that led to the present contribution. In this paper, we will focus on inhomogeneous Landau-type equations and establish the uniqueness within the classes of continuous or bounded solutions. More specifically, we study two types of equations. The first is a toy model \displaystyle\partial_{t}f+v\cdot\nabla_{x}f \displaystyle=\nabla_{v}\cdot\left(\rho(t,x)\left\langle v\right\rangle^{\beta}\nabla_{v}f\right),\qquad\rho(t,x)=\int_{\mathbb{R}^{3}}f(t,x,v)\,{\rm d}v, (1.1) and the second one is a viscous Landau-Coulomb equation \displaystyle\partial_{t}f+v\cdot\nabla_{x}f \displaystyle=\nabla_{v}\cdot\left(A[f]\nabla_{v}f-f\nabla_{v}a[f]\right)+\nu\Delta_{v}f, (1.2) with \nu>0 and A[f],\;a[f] defined as \displaystyle A[f](x,v,t) \displaystyle:=\frac{1}{8\pi}\int_{\mathbb{R}^{3}}\frac{\mathbb{P}(v-z)}{|v-z|}f(x,z,t)\;dz,\qquad\mathbb{P}(z):=\mathbb{I}d-\frac{z\otimes z}{|z|^{2}}, \displaystyle a[f](x,v,t) \displaystyle:=\frac{1}{4\pi}\int_{\mathbb{R}^{3}}\frac{1}{|v-z|}f(x,z,t)\;dz. The main results in this paper can be informally summarized as follows: Main Theorem . (a) Let \beta\leq 2 . Equation ( 1.1 ) has a unique solution that is continuous in (t,x,v) and has a fast enough polynomial decay tail in v . (b) Let \nu>0 be arbitrary. Equation ( 1.2 ) has a unique solution that is continuous in (t,x,v) with a fast enough polynomial decay tail in v . (c) Let \nu>0 be arbitrary and k_{0} large enough. Suppose \left\lVert\left\langle v\right\rangle^{k_{0}}f_{0}\right\rVert_{L^{\infty}_{x,v}} is small enough with its bound depending on \nu . Then ( 1.2 ) has a unique solution that is bounded in (t,x,v) with a sufficiently fast polynomial decay tail in v . The precise statements can be found in Theorem 3.2 , Theorem 4.2 and Theorem 4.3 . A common approach to proving the uniqueness of solutions to nonlinear diffusion equations is to identify a suitable function space in which some derivatives in x or v of the solution are bounded. Using ( 1.1 ) as an example, the difference between two solutions with same initial data, w:=f-g , satisfies \displaystyle\partial_{t}w+v\cdot\nabla_{x}w \displaystyle=\nabla_{v}\cdot\left(\rho_{f}(t,x)\left\langle v\right\rangle^{\beta}\nabla_{v}w\right)+\nabla_{v}\cdot\left(\rho_{w}(t,x)\left\langle v\right\rangle^{\beta}\nabla_{v}g\right), \displaystyle w|_{t=0} \displaystyle=0, where \rho_{f}(t,x)=\int_{\mathbb{R}^{3}}f(t,x,v)\,{\rm d}v and \rho_{w}(t,x)=\int_{\mathbb{R}^{3}}w(t,x,v)\,{\rm d}v . In a basic energy estimate, the last term would need a uniform bound in the variable x for \nabla_{v}g . Several uniqueness results for inhomogeneous kinetic equations present in the literature follow this approach. The most recent ones on the inhomogeneous Landau or non-cutoff Boltzmann equations are included in [ 2 , 7 , 15 , 16 , 17 ] . In [ 2 ] , it is shown that the non-cutoff Boltzmann equation with 2s -singularity, has unique solution in L^{\infty}_{t,x}H^{2s}_{v} . This corresponds to L^{\infty}_{t,x}H^{2}_{v} for the Landau equation. In [ 7 ] , uniqueness of solution to the non-cutoff Boltzmann and Landau equations is shown in the class of functions near the global Maxwellian in the space X=L^{1}_{k}L^{\infty}_{T}L^{2}_{v} composed of the Wiener algebra. The space X is stronger than C^{1/2}_{x}L^{\infty}_{T}L^{2}_{v} , where C^{1/2}_{x} denotes the Hölder space of order 1/2 in x . In recent work [ 15 , 16 ] kinetic Schauder estimates have been employed to prove the uniqueness of solutions to the non-cutoff Boltzmann and Landau equations if the initial data have an arbitrary amount of Hölder regularity in both x and v . These results have recently been improved by [ 17 ] to initial data with Hölder regularity in x and logarithmic modulus of continuity in v . In this paper, we take a different point of view and make use of negative Sobolev spaces. To illustrate the idea, we consider again the toy model ( 1.1 ) with \beta=0 without transport term: \displaystyle\partial_{t}f=\Delta_{v}\left(\rho(t,x)f\right). (1.3) We now use (-\Delta_{v})^{-1}f as a test function. This is effectively equivalent to perform H^{-1} estimates on \displaystyle\partial_{t}(-\Delta_{v})^{-1}f=\rho(t,x)f, for which derivative on f does not need to be bounded. Unfortunately, direct application of (-\Delta_{v})^{-1} to the kinetic transport term no longer works, as its commuter with v\cdot\nabla_{x} induces differentiation in x . To address this issue, we propose to use a pseudo-differential operator that plays the role of (-\Delta_{v})^{-1} and commutes with v\cdot\nabla_{x} in a controllable way. This operator will feature a time-averaged negative Sobolev derivative with intermediate time. Operators with a time-averaged positive Sobolev derivatives already appeared in studies of the well-posedness of nonlinear kinetic equations (see, for example, [ 20 ] ). The pseudo-differential operator we use here is inspired by the one introduced in [ 3 ] , used to prove the L^{2}_{x}H^{-s}_{v} to L^{2}_{x}L^{2}_{v} regularization estimates for the linearized Boltzmann equation. Let \widehat{w} be the Fourier transform of w in (x,v) . We define the Fourier multiplier operator with symbol {M} {\mathcal{M}}w:=({\mathcal{F}_{(x,v)\to(\xi,\eta)}})^{-1}({M}\widehat{w}), with \displaystyle{M}(t,T,\eta,\xi):=\frac{1}{\left(1+\delta\int_{t}^{T}\langle\xi+(t-\tau)\eta\rangle^{2}\;d\tau\right)^{\frac{1}{2}+\epsilon}}. (1.4) Here, \epsilon and \delta are positive constants that can be chosen as needed. More precise properties for ( 1.4 ) are summarized in Section 2 . However, let us analyze the main characteristics. The commutator of \mathcal{M} with v\cdot\nabla_{x} generates a backward diffusion in v , whose strength is proportional to the parameter \delta : \displaystyle[\mathcal{M},\partial_{t}+v\cdot\nabla_{x}]\sim-\delta(I-\Delta_{v})\mathcal{M}, (1.5) This commutator term will be controlled by the diffusion on the right-hand side of ( 1.1 ) or by viscosity term of ( 1.2 ). In the case of ( 1.1 ) the loss of weights (for \beta<0 ) can be overcome by building a negative velocity weight in \mathcal{M} : in Section 3 we introduce a dyadic decomposition in M to localize the velocity in each ring where |v|\in[2^{n},2^{n+2}] for each n\in\mathbb{N} . Unfortunately, the dyadic decomposition approach is not yet applicable to anisotropic diffusion; therefore, in the proof of the uniqueness of solution to ( 1.2 ) we use

Read the complete text