DeepAI AI Chat
Log In Sign Up

High-dimensional Bayesian inference via the Unadjusted Langevin Algorithm

by   Alain Durmus, et al.
Ecole Polytechnique

We consider in this paper the problem of sampling a high-dimensional probability distribution π having a density with respect to the Lebesgue measure on R^d, known up to a normalisation factor e^-U(x)/∫_R^de^-U(y)d y. Such problem naturally occurs for example in Bayesian inference and machine learning. Under the assumption that U is continuously differentiable, ∇ U is globally Lipschitz and U is strongly convex, we obtain non-asymptotic bounds for the convergence to stationarity in Wasserstein distance of order 2 and total variation distance of the sampling method based on the Euler discretization of the Langevin stochastic differential equation, for both constant and decreasing step sizes. The dependence on the dimension of the state space of the obtained bounds is studied to demonstrate the applicability of this method. The convergence of an appropriately weighted empirical measure is also investigated and bounds for the mean square error and exponential deviation inequality are reported for functions which are either Lipchitz continuous or measurable and bounded. An illustration to a Bayesian inference for binary regression is presented.


page 1

page 2

page 3

page 4


Higher Order Langevin Monte Carlo Algorithm

A new (unadjusted) Langevin Monte Carlo (LMC) algorithm with improved ra...

Discrete sticky couplings of functional autoregressive processes

In this paper, we provide bounds in Wasserstein and total variation dist...

Stability of doubly-intractable distributions

Doubly-intractable distributions appear naturally as posterior distribut...

Generalized bounds for active subspaces

The active subspace method, as a dimension reduction technique, can subs...

Non-asymptotic convergence bounds for modified tamed unadjusted Langevin algorithm in non-convex setting

We consider the problem of sampling from a high-dimensional target distr...

Bayesian inference of a non-local proliferation model

From a systems biology perspective the majority of cancer models, althou...

Overlap matrix concentration in optimal Bayesian inference

We consider models of Bayesian inference of signals with vectorial compo...