Search

We establish general conditions under which there exists uniform in time convergence between a stochastic process and its approximated system. These standardised conditions consist of a local in time estimate between the original and the approximated process as well as of a contraction property for one of the processes and a uniform control for the other one. Specifically, the results we present provide global in time error bounds for multiscale methods and numerical discretisations as well as uniform in time propagation of chaos bounds for mean-field particle systems. We provide a general method of proof which can be applied to many types of approximation. In all three scenarios, examples where the joint conditions are verified and uniform in time convergence is achieved are given., Comment: 25 pages, 1 figure

The problem of sampling according to the probability distribution minimizing a given free energy, using interacting particles unadjusted kinetic Langevin Monte Carlo, is addressed. In this setting, three sources of error arise, related to three parameters: the number of particles $N$, the discretization step size $h$, and the length of the trajectory $n$. The main result of the present work is a quantitative estimate of strong convergence in relative entropy, implying non-asymptotic bounds for the quadratic risk of Monte Carlo estimators for bounded observables. The numerical discretization scheme considered here is a second-order splitting method, as commonly used in practice. In addition to $N,h,n$, the dependency in the ambient dimension $d$ of the problem is also made explicit, under suitable conditions. The main results are proven under general conditions (regularity, moments, log-Sobolev inequality), for which tractable conditions are then provided. In particular, a Lyapunov analysis is conducted under more general conditions than previous works; the nonlinearity may not be small and it may not be convex along linear interpolations between measures., Comment: 35 pages

We study three kinetic Langevin samplers including the Euler discretization, the BU and the UBU splitting scheme. We provide contraction results in $L^1$-Wasserstein distance for non-convex potentials. These results are based on a carefully tailored distance function and an appropriate coupling construction. Additionally, the error in the $L^1$-Wasserstein distance between the true target measure and the invariant measure of the discretization scheme is bounded. To get an $\varepsilon$-accuracy in $L^1$-Wasserstein distance, we show complexity guarantees of order $\mathcal{O}(\sqrt{d}/\varepsilon)$ for the Euler scheme and $\mathcal{O}(d^{1/4}/\sqrt{\varepsilon})$ for the UBU scheme under appropriate regularity assumptions on the target measure. The results are applicable to interacting particle systems and provide bounds for sampling probability measures of mean-field type., Comment: 44 pages, 3 figures

Continuous-time Markov chains associated to finite-volume discretization schemes of Fokker-Planck equations are constructed. Sufficient conditions under which quantitative exponential decay in the $\phi$-entropy and Wasserstein distance are established, implying modified logarithmic Sobolev, Poincar\'e, and discrete Beckner inequalities. The results are not restricted to additive potentials and do not make use of discrete Bochner-type identities. The proof for the $\phi$-decay relies on a coupling technique due to Conforti, while the proof for the Wasserstein distance uses the path coupling method. Furthermore, exponential equilibration for discrete-time Markov chains is proved, based on an abstract discrete Bakry-Emery method and a path coupling.

We present a nonlinear (in the sense of McKean) generalization of Hamiltonian Monte Carlo (HMC) termed nonlinear HMC (nHMC) capable of sampling from nonlinear probability measures of mean-field type. When the underlying confinement potential is $K$-strongly convex and $L$-gradient Lipschitz, and the underlying interaction potential is gradient Lipschitz, nHMC can produce an $\varepsilon$-accurate approximation of a $d$-dimensional nonlinear probability measure in $L^1$-Wasserstein distance using $O((L/K) \log(1/\varepsilon))$ steps. Owing to a uniform-in-steps propagation of chaos phenomenon, and without further regularity assumptions, unadjusted HMC with randomized time integration for the corresponding particle approximation can achieve $\varepsilon$-accuracy in $L^1$-Wasserstein distance using $O( (L/K)^{5/3} (d/K)^{4/3} (1/\varepsilon)^{8/3} \log(1/\varepsilon) )$ gradient evaluations. These mixing/complexity upper bounds are a specific case of more general results developed in the paper for a larger class of non-logconcave, nonlinear probability measures of mean-field type., Comment: 40 pages, 2 figures

We study the long-time behaviour of both the classical second-order Langevin dynamics and the nonlinear second-order Langevin dynamics of McKean-Vlasov type. By a coupling approach, we establish global contraction in an $L^1$ Wasserstein distance with an explicit dimension-free rate for pairwise weak interactions. For external forces corresponding to a $\kappa$-strongly convex potential, a contraction rate of order $\mathcal{O}(\sqrt{\kappa})$ is obtained in certain cases. But the contraction result is not restricted to these forces. It rather includes multi-well potentials and non-gradient-type external forces as well as non-gradient-type repulsive and attractive interaction forces. The proof is based on a novel distance function which combines two contraction results for large and small distances and uses a coupling approach adjusted to the distance. By applying a componentwise adaptation of the coupling we provide uniform in time propagation of chaos bounds for the corresponding mean-field particle system., Comment: 34 pages, 2 figures

We develop a new approach to study the long time behaviour of solutions to nonlinear stochastic differential equations in the sense of McKean, as well as propagation of chaos for the corresponding mean-field particle system approximations. Our approach is based on a sticky coupling between two solutions to the equation. We show that the distance process between the two copies is dominated by a solution to a one-dimensional nonlinear stochastic differential equation with a sticky boundary at zero. This new class of equations is then analyzed carefully. In particular, we show that the dominating equation has a phase transition. In the regime where the Dirac measure at zero is the only invariant probability measure, we prove exponential convergence to equilibrium both for the one-dimensional equation, and for the original nonlinear SDE. Similarly, propagation of chaos is shown by a componentwise sticky coupling and comparison with a system of one dimensional nonlinear SDEs with sticky boundaries at zero. The approach applies to equations without confinement potential and to interaction terms that are not of gradient type., Comment: 46 pages

We present dimension-free convergence and discretization error bounds for the unadjusted Hamiltonian Monte Carlo algorithm applied to high-dimensional probability distributions of mean-field type. These bounds require the discretization step to be sufficiently small, but do not require strong convexity of either the unary or pairwise potential terms present in the mean-field model. To handle high dimensionality, our proof uses a particlewise coupling that is contractive in a complementary particlewise metric., Comment: 38 pages, 4 figures