Statistical deconvolution of the free Fokker-Planck equation at fixed time

DOI10.3150/21-BEJ1366MaRDI QIDQ2136995zbMATH OpenOpenAlexFDO

Authors Tien Dat Nguyen, Thanh Mai Pham Ngoc, V. Rivoirard, Viet Chi Tran, Mylène Maida

Publication date 16 May 2022

Published in Bernoulli (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/2006.11899, https://projecteuclid.org/journals/bernoulli/volume-28/issue-2/Statistical-deconvolution-of-the-free-Fokker-Planck-equation-at-fixed/10.3150/21-BEJ1366.full

zbMATH Keywords

Cauchy distributionDyson Brownian motionsFokker-Planck equationnonlinear partial differential equation

Mathematics Subject Classification ID

PDEs in connection with statistics (35Q62) Fokker-Planck equations (35Q84) Inverse problems for PDEs (35R30) Noncommutative probability and statistics (46L53) Free probability and free operator algebras (46L54) Random matrices (probabilistic aspects) (60B20) Brownian motion (60J65) Nonparametric estimation (62G05) Numerical methods for inverse problems for initial value and initial-boundary value problems involving PDEs (65M32)

Abstract: We are interested in reconstructing the initial condition of a non-linear partial differential equation (PDE), namely the Fokker-Planck equation, from the observation of a Dyson Brownian motion at a given time $t>0$. The Fokker-Planck equation describes the evolution of electrostatic repulsive particle systems, and can be seen as the large particle limit of correctly renormalized Dyson Brownian motions. The solution of the Fokker-Planck equation can be written as the free convolution of the initial condition and the semi-circular distribution. We propose a nonparametric estimator for the initial condition obtained by performing the free deconvolution via the subordination functions method. This statistical estimator is original as it involves the resolution of a fixed point equation, and a classical deconvolution by a Cauchy distribution. This is due to the fact that, in free probability, the analogue of the Fourier transform is the R-transform, related to the Cauchy transform. In past literature, there has been a focus on the estimation of the initial conditions of linear PDEs such as the heat equation, but to the best of our knowledge, this is the first time that the problem is tackled for a non-linear PDE. The convergence of the estimator is proved and the integrated mean square error is computed, providing rates of convergence similar to the ones known for non-parametric deconvolution methods. Finally, a simulation study illustrates the good performances of our estimator.

Review text

This paper is concerned with the reconstruction of the initial condition of a nonlinear partial differential equation, the so-called PDE or Fokker-Planck equation, from the observation of a Dyson Brownian motion at a given time $t>0$. Recall that the Fokker-Planck equation describes the evolution of electrostatic repulsive particle systems, its solution can be written as the free convolution of the initial condition and the semi-circular distribution. As a consequence of this, the authors are interested in estimating the initial condition of a nonlinear PDE from the observation of its solution at a specific point in time $t$. This is represented by the following equation: \[ \partial tp(t,x)=-\partial x\int R2Hp(t,x)p(t, x)dx \] Furthermore, a fundamental problem, according to the authors, letting the initial condition of a partial differential equation (PDE) be random, is its strong similarities with the heat equation, the well-known standard Brownian motion of the probabilistic interpretation. In this context, it is replaced by the free Brownian motion $(ht)t>0$, the operator-valued, and by a semi-circular distribution $\sigma t$ characterized by its density with respect to the Lebesgue measure. This problem was resolved introducing a nonparametric estimator for the initial condition obtained by performing the free deconvolution via the subordination functions method involving the resolution of a fixed point equation, and a classical deconvolution by a Cauchy distribution. $G\mu 0(7)=1t(wfp(z)-z)=G\mu t(wfp(z))$ (2) Consequently $|wfp(z-z|\leq\sqrt t$ (3) As a related result, they conducted an empirical simulation with the aim to assess the performance of their constructed statistical estimator based on a sample of $\delta n(t):=\delta n1(t),\dots, \delta nn(t)$ of non-ordered eigenvalues, or in other words, $n=400$ and $t=1$.

Recommendations

• Inverse problems in Fokker Planck equations
• Identification of the drift coeffic plank equation from the momer its stationary solution
• Statistical approach to dynamical inverse problems
• Fokker-Planck equations with terminal condition and related McKean probabilistic representation
• Kernel estimation of initial data of partial differential equations

Cites work

• A new approach to subordination results in free probability
• A stochastic particle method with random weights for the computation of statistical solutions of McKean-Vlasov equations
• A wavelet particle approximation for McKean-Vlasov and 2D-Navier-Stokes statistical solutions
• Adaptive quantile estimation in deconvolution with unknown error distribution
• Addition of certain non-commuting random variables
• An introduction to random matrices
• Anisotropic adaptive kernel deconvolution
• Density deconvolution under general assumptions on the distribution of measurement errors
• Free diffusions, free entropy and free Fisher information
• Functional deconvolution in a periodic setting: uniform case
• scientific article; zbMATH DE number 3868512 (Why is no real title available?)
• scientific article; zbMATH DE number 4211245 (Why is no real title available?)
• scientific article; zbMATH DE number 3473085 (Why is no real title available?)
• scientific article; zbMATH DE number 2147507 (Why is no real title available?)
• scientific article; zbMATH DE number 933353 (Why is no real title available?)
• Kinetic equilibration rates for granular media and related equations: entropy dissipation and mass transportation estimates
• Laplace deconvolution on the basis of time domain data and its application to dynamic contrast-enhanced imaging
• Mathematical Problems of Statistical Hydromechanics
• Minimax theory of estimation of linear functionals of the deconvolution density with or without sparsity
• On convergence rates equivalency and sampling strategies in functional deconvolution models
• On the free convolution with a semi-circular distribution
• Periodic nonlinear Schrödinger equation and invariant measures
• Random data Cauchy theory for supercritical wave equations I: Local theory
• Random data Cauchy theory for supercritical wave equations. II. A global existence result
• Random data wave equations
• Rates of convergence for nonparametric deconvolution
• Sharp Optimality in Density Deconvolution with Dominating Bias. I
• Spectral analysis of large dimensional random matrices
• Statistical deconvolution of the free Fokker-Planck equation at fixed time
• Statistical solutions of the Navier–Stokes equations on the phase space of vorticity and the inviscid limits
• Statistics of a flux in Burgers turbulence with one-sided Brownian initial data
• Subordination methods for free deconvolution
• The analogues of entropy and of Fisher's information measure in free probability theory. I
• Wavelet deconvolution
• Weak vorticity formulation of 2D Euler equations with white noise initial condition

Cited in

(2)

• Statistical deconvolution of the free Fokker-Planck equation at fixed time
• Polynomial rates via deconvolution for nonparametric estimation in McKean-Vlasov SDEs

This page was built for publication: Statistical deconvolution of the free Fokker-Planck equation at fixed time

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2136995)