Splitting for some classes of homeomorphic and coalescing stochastic flows

M.B. Vovchanskyi

DOI: https://doi.org/10.1051/ps/2024004

2024-04-05

ESAIM Probability and Statistics

Abstract:Authors: M.B. Vovchanskyi. ESAIM: Probability and Statistics Vol. 28 , page 75 Published online: 04/04/2024 Keywords: Splitting scheme ; stochastic flow ; stochastic differential equation.60H35 ; 60K35 ; 65C30.

statistics & probability

What problem does this paper attempt to address?

Splitting for some classes of homeomorphic and coalescing stochastic flows

M.B. Vovchanskyi

2024-03-08

Abstract:The splitting scheme (the Kato-Trotter formula) is applied to stochastic flows with common noise of the type introduced by Th.E.~Harris. The case of possibly coalescing flows with continuous infinitesimal covariance is considered and the weak convergence of the corresponding finite-dimensional motions is established. As applications, results for the convergence of the associated pushforward measures and dual flows are given. Similarities between splitting and the Euler-Maruyama scheme yield estimates of the speed of the convergence under additional regularity assumptions.

Probability
High Order Splitting Methods for SDEs Satisfying a Commutativity Condition

James M. Foster,Gonçalo dos Reis,Calum Strange

DOI: https://doi.org/10.1137/23m155147x

IF: 3.039

2024-02-16

SIAM Journal on Numerical Analysis

Abstract:SIAM Journal on Numerical Analysis, Volume 62, Issue 1, Page 500-532, February 2024. In this paper, we introduce a new simple approach to developing and establishing the convergence of splitting methods for a large class of stochastic differential equations (SDEs), including additive, diagonal, and scalar noise types. The central idea is to view the splitting method as a replacement of the driving signal of an SDE, namely, Brownian motion and time, with a piecewise linear path that yields a sequence of ODEs—which can be discretized to produce a numerical scheme. This new way of understanding splitting methods is inspired by, but does not use, rough path theory. We show that when the driving piecewise linear path matches certain iterated stochastic integrals of Brownian motion, then a high order splitting method can be obtained. We propose a general proof methodology for establishing the strong convergence of these approximations that is akin to the general framework of Milstein and Tretyakov. That is, once local error estimates are obtained for the splitting method, then a global rate of convergence follows. This approach can then be readily applied in future research on SDE splitting methods. By incorporating recently developed approximations for iterated integrals of Brownian motion into these piecewise linear paths, we propose several high order splitting methods for SDEs satisfying a certain commutativity condition. In our experiments, which include the Cox–Ingersoll–Ross model and additive noise SDEs (noisy anharmonic oscillator, stochastic FitzHugh–Nagumo model, underdamped Langevin dynamics), the new splitting methods exhibit convergence rates of [math] and outperform schemes previously proposed in the literature.

mathematics, applied
Isotropic Stochastic Flow of Homeomorphisms on Sd for the Critical Sobolev Exponent

SZ Fang,TS Zhang

DOI: https://doi.org/10.1016/j.matpur.2005.10.012

IF: 2.282

2006-01-01

Journal de Mathématiques Pures et Appliquées

Abstract:In this work, we shall deal with the critical Sobolev isotropic Brownian flows on the sphere S-d. Based on previous works by O. Raimond and LeJan and Raimond (see [O. Raimond, Ann. Inst. H. Poincare 35 (1999) 313-354] and [Y LeJan, O. Raimond, Ann. of Prob. 30 (2002) 826-873], we prove that the associated flows are flows of homeomorphisms. (c) 2005 Elsevier SAS. All rights reserved.
An almost symmetric Strang splitting scheme for nonlinear evolution equations

Lukas Einkemmer,Alexander Ostermann

DOI: https://doi.org/10.1016/j.camwa.2014.02.027

2013-09-17

Abstract:In this paper we consider splitting methods for the time integration of parabolic and certain classes of hyperbolic partial differential equations, where one partial flow can not be computed exactly. Instead, we use a numerical approximation based on the linearization of the vector field. This is of interest in applications as it allows us to apply splitting methods to a wider class of problems from the sciences. However, in the situation described the classic Strang splitting scheme, while still a method of second order, is not longer symmetric. This, in turn, implies that the construction of higher order methods by composition is limited to order three only. To remedy this situation, based on previous work in the context of ordinary differential equations, we construct a class of Strang splitting schemes that are symmetric up to a desired order. We show rigorously that, under suitable assumptions on the nonlinearity, these methods are of second order and can then be used to construct higher order methods by composition. In addition, we illustrate the theoretical results by conducting numerical experiments for the Brusselator system and the KdV equation.

Numerical Analysis
A randomized operator splitting scheme inspired by stochastic optimization methods

Monika Eisenmann,Tony Stillfjord

DOI: https://doi.org/10.1007/s00211-024-01396-w

IF: 2.5

2024-02-27

Numerische Mathematik

Abstract:In this paper, we combine the operator splitting methodology for abstract evolution equations with that of stochastic methods for large-scale optimization problems. The combination results in a randomized splitting scheme, which in a given time step does not necessarily use all the parts of the split operator. This is in contrast to deterministic splitting schemes which always use every part at least once, and often several times. As a result, the computational cost can be significantly decreased in comparison to such methods. We rigorously define a randomized operator splitting scheme in an abstract setting and provide an error analysis where we prove that the temporal convergence order of the scheme is at least 1/2. We illustrate the theory by numerical experiments on both linear and quasilinear diffusion problems, using a randomized domain decomposition approach. We conclude that choosing the randomization in certain ways may improve the order to 1. This is as accurate as applying e.g. backward (implicit) Euler to the full problem, without splitting.

mathematics, applied
Solving Backward Doubly Stochastic Differential Equations through Splitting Schemes

Feng Bao,Yanzhao Cao,He Zhang

DOI: https://doi.org/10.48550/arXiv.2103.08632

2021-03-15

Numerical Analysis

Abstract:A splitting scheme for backward doubly stochastic differential equations is proposed. The main idea is to decompose a backward doubly stochastic differential equation into a backward stochastic differential equation and a stochastic differential equation. The backward stochastic differential equation and the stochastic differential equation are then approximated by first order finite difference schemes, which results in a first order scheme for the backward doubly stochastic differential equation. Numerical experiments are conducted to illustrate the convergence rate of the proposed scheme.
Splitting Methods for differential equations

Sergio Blanes,Fernando Casas,Ander Murua

2024-05-07

Abstract:This overview is devoted to splitting methods, a class of numerical integrators intended for differential equations that can be subdivided into different problems easier to solve than the original system. Closely connected with this class of integrators are composition methods, in which one or several low-order schemes are composed to construct higher-order numerical approximations to the exact solution. We analyze in detail the order conditions that have to be satisfied by these classes of methods to achieve a given order, and provide some insight about their qualitative properties in connection with geometric numerical integration and the treatment of highly oscillatory problems. Since splitting methods have received considerable attention in the realm of partial differential equations, we also cover this subject in the present survey, with special attention to parabolic equations and their problems. An exhaustive list of methods of different orders is collected and tested on simple examples. Finally, some applications of splitting methods in different areas, ranging from celestial mechanics to statistics, are also provided.

Numerical Analysis
A new class of splitting methods that preserve ergodicity and exponential integrability for stochastic Langevin equation

Chuchu Chen,Tonghe Dang,Jialin Hong,Fengshan Zhang

2024-10-28

Abstract:In this paper, we propose a new class of splitting methods to solve the stochastic Langevin equation, which can simultaneously preserve the ergodicity and exponential integrability of the original equation. The central idea is to extract a stochastic subsystem that possesses the strict dissipation from the original equation, which is inspired by the inheritance of the Lyapunov structure for obtaining the ergodicity. We prove that the exponential moment of the numerical solution is bounded, thus validating the exponential integrability of the proposed methods. Further, we show that under moderate verifiable conditions, the methods have the first-order convergence in both strong and weak senses, and we present several concrete splitting schemes based on the methods. The splitting strategy of methods can be readily extended to construct conformal symplectic methods and high-order methods that preserve both the ergodicity and the exponential integrability, as demonstrated in numerical experiments. Our numerical experiments also show that the proposed methods have good performance in the long-time simulation.

Numerical Analysis,Probability
On the splitting-up method for rough (partial) differential equations

Peter Friz,Harald Oberhauser

DOI: https://doi.org/10.48550/arXiv.1008.0513

2010-08-03

Abstract:This article introduces the splitting method to systems responding to rough paths as external stimuli. The focus is on nonlinear partial differential equations with rough noise but we also cover rough differential equations. Applications to stochastic partial differential equations arising in control theory and nonlinear filtering are given.

Probability,Analysis of PDEs
Transformation of Stochastic Recursions and Critical Phenomena in the Analysis of a Class of Mean Flow Equations

Radu Dascaliuc,Tuan N. Pham,Enrique Thomann,Edward C. Waymire

2024-11-01

Abstract:This paper investigates, via methods from the theory of probability on trees, critical phenomena in stochastic cascade models of Yule type, and applies these methods to the problem of uniqueness and nonuniqueness of solutions of particular mean flow equations. For linear equations, Feller's classic discovery of the relationship between critical stochastic explosion phenomena for Markov processes and uniqueness and nonuniqueness of solutions to the associated Kolmogorov equation partly illustrates the spirit of the present paper. New methods are introduced to explore the effects that stochastic critical phenomena, such as stochastic explosion, hyperexplosion, t-leaf percolation, have on their corresponding mean flow equations. These methods are illustrated in the context of two non-local differential equations that are related by linearization, the pantograph and the $\alpha$-Riccati equations. In particular, stochastic recursions, stochastic Picard iterations and transformation of solutions processes associated with these equations, are used to establish a `one-to-many principle' for their solutions. The development also includes an application of Feller's classical theory of jump Markov processes to a certain class of pantograph equations that appears to be new.

Probability,Analysis of PDEs
Random Splitting of Point Vortex Flows

Andrea Agazzi,Francesco Grotto,Jonathan C. Mattingly

2023-11-27

Abstract:We consider a stochastic version of the point vortex system, in which the fluid velocity advects single vortices intermittently for small random times. Such system converges to the deterministic point vortex dynamics as the rate at which single components of the vector field are randomly switched diverges, and therefore it provides an alternative discretization of 2D Euler equations. The random vortex system we introduce preserves microcanonical statistical ensembles of the point vortex system, hence constituting a simpler alternative to the latter in the statistical mechanics approach to 2D turbulence.

Probability,Mathematical Physics,Dynamical Systems
Single-Source Unsplittable Flows in Planar Graphs

Vera Traub,Laura Vargas Koch,Rico Zenklusen

2023-08-05

Abstract:The single-source unsplittable flow (SSUF) problem asks to send flow from a common source to different terminals with unrelated demands, each terminal being served through a single path. One of the most heavily studied SSUF objectives is to minimize the violation of some given arc capacities. A seminal result of Dinitz, Garg, and Goemans showed that, whenever a fractional flow exists respecting the capacities, then there is an unsplittable one violating the capacities by at most the maximum demand. Goemans conjectured a very natural cost version of the same result, where the unsplittable flow is required to be no more expensive than the fractional one. This intriguing conjecture remains open. More so, there are arguably no non-trivial graph classes for which it is known to hold. We show that a slight weakening of it (with at most twice as large violations) holds for planar graphs. Our result is based on a connection to a highly structured discrepancy problem, whose repeated resolution allows us to successively reduce the number of paths used for each terminal, until we obtain an unsplittable flow. Moreover, our techniques also extend to simultaneous upper and lower bounds on the flow values. This also affirmatively answers a conjecture of Morell and Skutella for planar SSUF.

Data Structures and Algorithms
Remarks on the separation of Navier-Stokes flows

Zachary Bradshaw

DOI: https://doi.org/10.48550/arXiv.2312.11353

2023-12-19

Abstract:Recently, strong evidence has accumulated that some solutions to the Navier-Stokes equations in physically meaningful classes are not unique. The primary purpose of this paper is to establish necessary properties for the error of hypothetical non-unique Navier-Stokes flows under conditions motivated by the scaling of the equations. Our first set of results show that some scales are necessarily active -- comparable in norm to the full error -- as solutions separate. `Scale' is interpreted in several ways, namely via algebraic bounds, the Fourier transform and discrete volume elements. These results include a new type of uniqueness criteria which is stated in terms of the error. The second result is a conditional predictability criteria for the separation of small perturbations. An implication is that the error necessarily activates at larger scales as flows de-correlate. The last result says that the error of the hypothetical non-unique Leray-Hopf solutions of Jia and Sverak locally grows in a self-similar fashion. Consequently, within the Leray-Hopf class, energy can de-correlate at a rate which is faster than linear. This contrasts numerical work on predictability which identifies a linear rate. This discrepancy can likely be explained by the fact that non-uniqueness can be viewed as a perturbation of a singular flow.

Analysis of PDEs
Numerical computations of split Bregman method for fourth order total variation flow

Yoshikazu Giga,Yuki Ueda

DOI: https://doi.org/10.1016/j.jcp.2019.109114

2019-06-11

Abstract:The split Bregman framework for Osher-Solé-Vese (OSV) model and fourth order total variation flow are studied. We discretize the problem by piecewise constant function and compute $\nabla(-\Delta_{\mathrm{av}})^{-1}$ approximately and exactly. Furthermore, we provide a new shrinkage operator for Spohn's fourth order model. Numerical experiments are demonstrated for fourth order problems under periodic boundary condition.

Numerical Analysis
Splitting schemes for hyperbolic heat conduction equation

Petr N. Vabishchevich

DOI: https://doi.org/10.1007/s10543-013-0423-7

IF: 1.856

2013-03-08

BIT Numerical Mathematics

Abstract:Rapid processes of heat transfer are not described by the standard heat conduction equation. To take into account a finite velocity of heat transfer, we use the hyperbolic model of heat conduction, which is connected with the relaxation of heat fluxes. In this case, the mathematical model is based on a hyperbolic equation of second order or a system of equations for the temperature and heat fluxes. In this paper we construct for the hyperbolic heat conduction equation the additive schemes of splitting with respect to directions. Unconditional stability of locally one-dimensional splitting schemes is established. New splitting schemes are proposed and studied for a system of equations written in terms of the temperature and heat fluxes.

mathematics, applied,computer science, software engineering
New Splitting Methods for Convection-Dominated Diffusion Problems and Navier-Stokes Equations

Feng Shi,Guoping Liang,Yubo Zhao,Jun Zou

DOI: https://doi.org/10.4208/cicp.031013.030614a

IF: 3.791

2014-01-01

Communications in Computational Physics

Abstract:We present a new splitting method for time-dependent convention-dominated diffusion problems. The original convention diffusion system is split into two sub-systems: a pure convection system and a diffusion system. At each time step, a convection problem and a diffusion problem are solved successively. A few important features of the scheme lie in the facts that the convection subproblem is solved explicitly and multistep techniques can be used to essentially enlarge the stability region so that the resulting scheme behaves like an unconditionally stable scheme; while the diffusion subproblem is always self-adjoint and coercive so that they can be solved efficiently using many existing optimal preconditioned iterative solvers. The scheme can be extended for solving the Navier-Stokes equations, where the nonlinearity is resolved by a linear explicit multistep scheme at the convection step, while only a generalized Stokes problem is needed to solve at the diffusion step and the major stiffness matrix stays invariant in the time marching process. Numerical simulations are presented to demonstrate the stability, convergence and performance of the single-step and multistep variants of the new scheme.
A New Splitting Method for Time-dependent Convection-dominated Diffusion Problems

Feng Shi,Guoping Liang,Yubo Zhao,Jun Zou

DOI: https://doi.org/10.48550/arxiv.1302.4032

2013-01-01

Abstract:We present a new splitting method for time-dependent convection-dominated diffusion problems. The original convection diffusion system is split into two sub-systems: a pure convection system and a diffusion system. At each time step, a convection problem and a diffusion problem are solved successively. The scheme has the following nice features: the convection subproblem is solved explicitly and a multistep technique is introduced to essentially enlarge the stability region so that the resulting scheme behaves like an unconditionally stable scheme; the diffusion subproblem is always self-adjoint and coercive so that it can be solved efficiently using many existing optimal preconditioned iterative solvers. The scheme is then extended for Navier-Stokes equations, where the nonlinear convection is resolved by a linear explicit multistep scheme at the convection step, and only a generalized Stokes problem is needed to solve at the diffusion step with the resulting stiffness matrix being invariant in the time marching process. The new schemes are all free from tuning some stabilization parameters for the convection-dominated diffusion problems. Numerical simulations are presented to demonstrate the stability, convergence and performance of the single-step and multistep variants of the new scheme.
Numerical Transport Process of Splitting Kinetic Schemes in the Navier-Stokes-Fourier Limit

Yajun Zhu,Chengwen Zhong,Kun Xu

DOI: https://doi.org/10.1080/10618562.2021.2023737

IF: 1.443

2021-01-01

International Journal of Computational Fluid Dynamics

Abstract:The Boltzmann equation is the fundamental governing equation in rarefied gas dynamics. Due to the complexity of Boltzmann collision term, operator splitting treatment is commonly adopted, where the Boltzmann equation is split into a convection equation for particles' free transport and an ordinary differential equation for particles' collision. However, this split treatment will introduce numerical error proportional to the time step, which may contaminate the physical solution in the near continuum regime. Therefore, for a multiscale kinetic method, the asymptotic preserving property to obtain the Navier-Stokes-Fourier (NSF) solution in the hydrodynamic limit is very important. In this paper, we analyse the effective relaxation time from different evolution processes of several kinetic schemes and investigate their capabilities to recover the NSF solution. The general requirement on a splitting kinetic method for the NSF solution has been presented. Numerical validation has been carried out, which shows good agreement with the theoretical analysis.
Control theory and splitting methods

Karine Beauchard,Adrien Laurent,Frédéric Marbach

2024-07-02

Abstract:Our goal is to highlight some of the deep links between numerical splitting methods and control theory. We consider evolution equations of the form $\dot{x} = f_0(x) + f_1(x)$, where $f_0$ encodes a non-reversible dynamic, so that one is interested in schemes only involving forward flows of $f_0$. In this context, a splitting method can be interpreted as a trajectory of the control-affine system $\dot{x}(t)=f_0(x(t))+u(t)f_1(x(t))$, associated with a control~$u$ which is a finite sum of Dirac masses. The general goal is then to find a control such that the flow of $f_0 + u(t) f_1$ is as close as possible to the flow of $f_0+f_1$. Using this interpretation and classical tools from control theory, we revisit well-known results concerning numerical splitting methods, and we prove a handful of new ones, with an emphasis on splittings with additional positivity conditions on the coefficients. First, we show that there exist numerical schemes of any arbitrary order involving only forward flows of $f_0$ if one allows complex coefficients for the flows of $f_1$. Equivalently, for complex-valued controls, we prove that the Lie algebra rank condition is equivalent to the small-time local controllability of a system. Second, for real-valued coefficients, we show that the well-known order restrictions are linked with so-called "bad" Lie brackets from control theory, which are known to yield obstructions to small-time local controllability. We use our recent basis of the free Lie algebra to precisely identify the conditions under which high-order methods exist.

Numerical Analysis,Optimization and Control
Variational closures for composite homogenised fluid flows

Theo Diamantakis,Ruiao Hu

2024-09-16

Abstract:The Stochastic Advection by Lie Transport is a variational formulation of stochastic fluid dynamics introduced to model the effects of unresolved scales, whilst preserving the geometric structure of ideal fluid flows. In this work, we show that the SALT equations can arise from the decomposition of the fluid flow map into its mean and fluctuating components. The fluctuating component is realised as a prescribed stochastic diffeomorphism that introduces stochastic transport into the system and we construct it using homogenisation theory. The dynamics of the mean component are derived from a variational principle utilising particular forms of variations that preserve the composite structure of the flow. Using a new variational principle, we show that SALT equations can arise from random Lagrangians and are equivalent to random coefficient PDEs. We also demonstrate how to modify the composite flow and the associated variational principle to derive models inspired by the Lagrangian Averaged Euler-Poincare (LAEP) theory.

Mathematical Physics,Dynamical Systems,Probability,Fluid Dynamics

Splitting for some classes of homeomorphic and coalescing stochastic flows

Splitting for some classes of homeomorphic and coalescing stochastic flows

High Order Splitting Methods for SDEs Satisfying a Commutativity Condition

Isotropic Stochastic Flow of Homeomorphisms on Sd for the Critical Sobolev Exponent

An almost symmetric Strang splitting scheme for nonlinear evolution equations

A randomized operator splitting scheme inspired by stochastic optimization methods

Solving Backward Doubly Stochastic Differential Equations through Splitting Schemes

Splitting Methods for differential equations

A new class of splitting methods that preserve ergodicity and exponential integrability for stochastic Langevin equation

On the splitting-up method for rough (partial) differential equations

Transformation of Stochastic Recursions and Critical Phenomena in the Analysis of a Class of Mean Flow Equations

Random Splitting of Point Vortex Flows

Single-Source Unsplittable Flows in Planar Graphs

Remarks on the separation of Navier-Stokes flows

Numerical computations of split Bregman method for fourth order total variation flow

Splitting schemes for hyperbolic heat conduction equation

New Splitting Methods for Convection-Dominated Diffusion Problems and Navier-Stokes Equations

A New Splitting Method for Time-dependent Convection-dominated Diffusion Problems

Numerical Transport Process of Splitting Kinetic Schemes in the Navier-Stokes-Fourier Limit

Control theory and splitting methods

Variational closures for composite homogenised fluid flows