Answers are grounded in this paper and cite their sources.
We consider sampling from a Gibbs distribution by evolving finitely many particles. We propose a preconditioned version of a recently proposed noise-free sampling method, governed by approximating the score function with the numerically tractable score of a regularized Wasserstein proximal operator. This is derived by a Cole–Hopf transformation on coupled anisotropic heat equations, yielding a kernel formulation for the preconditioned regularized Wasserstein proximal. The diffusion component of the proposed method is also interpreted as a modified self-attention block, as in transformer architectures. For quadratic potentials, we provide a discrete-time nonasymptotic convergence analysis and explicitly characterize the bias, which is dependent on regularization and independent of step-size. Experiments demonstrate acceleration and particle-level stability on various log-concave and non-log-concave toy examples, as well as for Bayesian total-variation regularized image deconvolution. We further show competitive/better performance on nonconvex Bayesian neural network training when utilizing variable preconditioning matrices.
No takes yet. Share an insight, caveat, or question.
Tan et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: