Paperlane

no. 12 · Tuesday 10 November 2026

Preconditioned Implicit Midpoint Langevin Sampling for Non-Smooth Bayesian Imaging

Sichen Wang and Zhipeng Lu

math.NA · via cs.CV · 40 pp

Abstract

We introduce P-IMLA, a preconditioned implicit-midpoint Langevin sampler for high-dimensional log-concave targets with non-smooth regularization. The obstruction is that the large steps enabled by matrix preconditioning are precisely where preconditioned Euler-Maruyama inflates stationary variance. Implicit midpoint removes this surplus via a Cayley-transform identity, making P-IMLA Gaussian-exact at any step size. Beyond Gaussians, whitening transfers the sharp IMLA contraction rate to $κ_{\mathrm{eff}}$, and backward-error analysis gives drift-only rather than Euler-type diffusion bias under explicit smoothness assumptions. An $M$-norm Moreau envelope yields a proximal non-smooth variant with a three-term error budget. Controlled Gaussian and TV-regularized imaging experiments confirm the predicted separation: the unpreconditioned midpoint scheme retains long transients as $κ_f$ grows, whereas P-IMLA remains $κ_{\mathrm{eff}}$-governed and near variance-exact.

comments
Accepted at NeurIPS 2026. 40 pages, 19 figures
fields
Numerical Analysis, Optimization and Control
pdf ↗ page arXiv:2611.09980 ↗