Importance corrected neural JKO sampling

J Hertrich, R Gruhlke - arXiv preprint arXiv:2407.20444, 2024 - arxiv.org
In order to sample from an unnormalized probability density function, we propose to
combine continuous normalizing flows (CNFs) with rejection-resampling steps based on …

Stein transport for Bayesian inference

N Nüsken - arXiv preprint arXiv:2409.01464, 2024 - arxiv.org
We introduce $\textit {Stein transport} $, a novel methodology for Bayesian inference
designed to efficiently push an ensemble of particles along a predefined curve of tempered …

Efficient, multimodal, and derivative-free bayesian inference with Fisher–Rao gradient flows

Y Chen, DZ Huang, J Huang, S Reich… - Inverse Problems, 2024 - iopscience.iop.org
In this paper, we study efficient approximate sampling for probability distributions known up
to normalization constants. We specifically focus on a problem class arising in Bayesian …

[PDF][PDF] A connection between tempering and entropic mirror descent

N Chopin, FR Crucinio, A Korba - arXiv preprint arXiv:2310.11914, 2023 - arxiv.org
This paper explores the connections between tempering (for Sequential Monte Carlo; SMC)
and entropic mirror descent to sample from a target probability distribution whose …

Wasserstein gradient flows of MMD functionals with distance kernel and Cauchy problems on quantile functions

R Duong, V Stein, R Beinert, J Hertrich… - arXiv preprint arXiv …, 2024 - arxiv.org
We give a comprehensive description of Wasserstein gradient flows of maximum mean
discrepancy (MMD) functionals $\mathcal F_\nu:=\text {MMD} _K^ 2 (\cdot,\nu) $ towards …

Sequential-in-time training of nonlinear parametrizations for solving time-dependent partial differential equations

H Zhang, Y Chen, E Vanden-Eijnden… - arXiv preprint arXiv …, 2024 - arxiv.org
Sequential-in-time methods solve a sequence of training problems to fit nonlinear
parametrizations such as neural networks to approximate solution trajectories of partial …

Fisher-rao gradient flow: geodesic convexity and functional inequalities

JA Carrillo, Y Chen, DZ Huang, J Huang… - arXiv preprint arXiv …, 2024 - arxiv.org
The dynamics of probability density functions has been extensively studied in science and
engineering to understand physical phenomena and facilitate algorithmic design. Of …

Ensemble-Based Annealed Importance Sampling

H Chen, L Ying - arXiv preprint arXiv:2401.15645, 2024 - arxiv.org
Sampling from a multimodal distribution is a fundamental and challenging problem in
computational science and statistics. Among various approaches proposed for this task, one …

Connections between sequential Bayesian inference and evolutionary dynamics

S Pathiraja, P Wacker - arXiv preprint arXiv:2411.16366, 2024 - arxiv.org
It has long been posited that there is a connection between the dynamical equations
describing evolutionary processes in biology and sequential Bayesian learning methods …

Wasserstein Gradient Flows of MMD Functionals with Distance Kernels under Sobolev Regularization

R Duong, N Rux, V Stein, G Steidl - arXiv preprint arXiv:2411.09848, 2024 - arxiv.org
We consider Wasserstein gradient flows of maximum mean discrepancy (MMD) functionals
$\text {MMD} _K^ 2 (\cdot,\nu) $ for positive and negative distance kernels $ K (x, y):=\pm| xy …