Time-Inhomogeneous Preconditioned Langevin Dynamics

arXiv:2605.0609145.1
Predicted impact top 39% in ST · last 90 daysOriginality Incremental advance
AI Analysis

For practitioners of Bayesian inference and sampling, this method resolves a known trade-off in preconditioned Langevin dynamics, but the experimental validation is limited to low-dimensional or simple tasks.

The paper proposes TIPreL, a time- and position-dependent preconditioner for Langevin dynamics that simultaneously addresses global mode coverage and local mode exploration, proving convergence in Wasserstein-2 distance and demonstrating efficiency on ill-posed and Bayesian logistic regression tasks.

Langevin sampling from distributions of the form $p(x) \propto \exp(-Ψ(x))$ faces two major challenges: (global) mode coverage and (local) mode exploration. The first challenge is particularly relevant for multi-modal distributions with disjoint modes, whereas the second arises when the potential $Ψ$ exhibits diverse and ill-conditioned local mode geometry. To address these challenges, a common approach is to precondition Langevin dynamics with problem-specific information, such as the sample covariance or the local curvature of $Ψ$. However, existing preconditioner choices inherently involve a trade-off between global mode coverage and local mode exploration, and no prior method resolves both simultaneously. To overcome this limitation, we propose the TIPreL, which introduces a time- and position-dependent preconditioner. This design effectively addresses both challenges mentioned above within a single framework. We establish convergence of the resulting dynamics in the Wasserstein-2 distance both in continuous time and for a tamed Euler discretization. In particular, our analysis extends the existing state of the art by proving convergence under time- and space-dependent diffusion coefficients, and only locally Lipschitz drifts, which has not been covered by prior work. Finally, we experimentally compare TIPreL with competing preconditioning schemes on a two-dimensional, severely ill-posed example and on a Bayesian logistic regression task in higher dimensions, confirming the efficiency of the proposed method.

Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes