Empirical approximation to invariant measures for McKean-Vlasov processes: mean-field interaction vs self-interaction

DOI10.3150/22-BEJ1550MaRDI QIDQ6103256zbMATH OpenFDO

Authors Kai Du, Yifan Jiang, Jinfeng Li

Publication date 2 June 2023

Published in Bernoulli (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/2112.14112, https://projecteuclid.org/journals/bernoulli/volume-29/issue-3/Empirical-approximation-to-invariant-measures-for-McKeanVlasov-processes-Mean/10.3150/22-BEJ1550.full

zbMATH Keywords

approximating measuresempirical measuresinvariant measuresMcKean-Vlasov processesWasserstein distances

Mathematics Subject Classification ID

Probability measures on topological spaces (60B05) Central limit and other weak theorems (60F05) Limit theorems in probability theory (60F99) Stochastic ordinary differential equations (aspects of stochastic analysis) (60H10) Computational methods for stochastic equations (aspects of stochastic analysis) (60H35) Diffusion processes (60J60)

Abstract: This paper proves that, under a monotonicity condition, the invariant probability measure of a McKean--Vlasov process can be approximated by weighted empirical measures of some processes including itself. These processes are described by distribution dependent or empirical measure dependent stochastic differential equations constructed from the equation for the McKean--Vlasov process. Convergence of empirical measures is characterized by upper bound estimates for their Wasserstein distance to the invariant measure. The theoretical results are demonstrated via a mean-field Ornstein--Uhlenbeck process.

Review text

It is well known that the McKean-Vlasov process is a stochastic process described by a stochastic differential equation (SDE), whose coefficients depend on the distribution of the solution. Therefore it is not a Markov process in the usual sense. The equation is often referred as the McKean-Vlasov SDE or the mean-field SDE, deduced from the ``propagation of chaos of particle systems with mean-field interaction. In this paper, the authors consider the following problem for McKean-Vlasov processes: whether and how the invariant measure of a McKean-Vlasov process can be approximated by the empirical measures of the process itself or some other processes. It is proved that, under a monotonicity condition, the invariant probability measure of a McKean-Vlasov process can be approximated by weighted empirical measures of some processes including itself. The main contribution of the paper is that the authors construct a theoretical foundation for designing online algorithms to evaluate invariant measures for a class of McKean-Vlasov processes. The empirical approximation overcomes two shortages of the classical propagation of chaos. First, an unbiased approximation as time goes to infinity is provided, and even one trajectory is enough for computing the approximating measures. Second, the algorithm allows to make full use of the history information. To some extent, the authors compute infinite sample paths by admitting the equivalence between the temporal and spatial average. In addition, the advantages of the authors' methods, comparing to the ``propagation of chaos, are presented, as well as a comparison with previous results in this area and with similar algorithms.

Recommendations

• On ergodic measures for McKean-Vlasov stochastic equations
• Long-time behaviors of mean-field interacting particle systems related to McKean-Vlasov equations
• A law of large numbers for interacting diffusions via a mild formulation
• Rate of convergence of a particle method to the solution of the McKean-Vlasov equation
• Existence of invariant probability measures for functional McKean-Vlasov SDEs

Cites work

• A CLASS OF MARKOV PROCESSES ASSOCIATED WITH NONLINEAR PARABOLIC EQUATIONS
• A concise course on stochastic partial differential equations
• A PDE approach to a 2-dimensional matching problem
• A survey of random processes with reinforcement
• Asymptotic behavior of Brownian polymers
• Convergence rates for empirical measures of Markov chains in dual and Wasserstein distances
• Distribution dependent SDEs for Landau type equations
• Ergodicity for Infinite Dimensional Systems
• Ergodicity of self-attracting motion
• Existence and uniqueness theorems for solutions of McKean-Vlasov stochastic equations
• Existence of invariant probability measures for functional McKean-Vlasov SDEs
• Exponential convergence in entropy and Wasserstein for McKean-Vlasov SDEs
• scientific article; zbMATH DE number 3121490 (Why is no real title available?)
• scientific article; zbMATH DE number 4211245 (Why is no real title available?)
• Kinetic equilibration rates for granular media and related equations: entropy dissipation and mass transportation estimates
• Long-time behaviors of mean-field interacting particle systems related to McKean-Vlasov equations
• McKean-Vlasov SDEs under measure dependent Lyapunov conditions
• Mean field games
• Mean rates of convergence of empirical measures in the Wasserstein metric
• Mean-field stochastic differential equations and associated PDEs
• New particle representations for ergodic McKean-Vlasov SDEs
• Nonlinear Fokker-Planck equations for probability measures on path space and path-distribution dependent sdes
• Nonlinear self-stabilizing processes. II: Convergence to invariant probability
• On convergence to stationary distributions for solutions of nonlinear Fokker-Planck-Kolmogorov equations
• On ergodic measures for McKean-Vlasov stochastic equations
• On invariant measures of nonlinear Markov processes
• On the mean speed of convergence of empirical and occupation measures in Wasserstein distance
• On the rate of convergence in Wasserstein distance of the empirical measure
• Optimal Transport
• Probabilistic theory of mean field games with applications II. Mean field games with common noise and master equations
• Propagation of chaos for stochastic spatially structured neuronal networks with delay driven by jump diffusions
• Propagation of chaos in neural fields
• Quantitative Harris-type theorems for diffusions and McKean-Vlasov processes
• Recursive computation of the invariant distribution of a diffusion
• Self-interacting diffusions.
• Some limit theorems for empirical processes (with discussion)
• Wasserstein convergence rate for empirical measures on noncompact manifolds

Cited in

(14)

• Barriers of the McKean-Vlasov energy via a mountain pass theorem in the space of probability measures
• Online parameter estimation for the McKean-Vlasov stochastic differential equation
• Occupied processes: going with the flow
• Sharp L^q-convergence rate in p-Wasserstein distance for empirical measures of diffusion processes
• Amplitude equation for N-SPDEs with cubic nonlinearity in mean field limit
• Self-interacting approximation to McKean-Vlasov long-time limit: a Markov chain Monte Carlo method
• Empirical approximation to invariant measures of non-degenerate McKean-Vlasov dynamics
• Brownian particles controlled by their occupation measure
• Self-interacting CBO: existence, uniqueness, and long-time convergence
• Convergence in Wasserstein distance for empirical measures of non-symmetric subordinated diffusion processes
• Euler-type approximation for the invariant measure: an abstract framework
• Unbiased approximations for stationary distributions of McKean-Vlasov SDEs
• Asymptotics in Wasserstein distance for empirical measures of Markov processes
• Solving McKean-Vlasov stochastic differential equations by a deep learning particle method

This page was built for publication: Empirical approximation to invariant measures for McKean-Vlasov processes: mean-field interaction vs self-interaction

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6103256)