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 …
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
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 …
to normalization constants. We specifically focus on a problem class arising in Bayesian …
Sampling in Unit Time with Kernel Fisher-Rao Flow
We introduce a new mean-field ODE and corresponding interacting particle systems for
sampling from an unnormalized target density or Bayesian posterior. The interacting particle …
sampling from an unnormalized target density or Bayesian posterior. The interacting particle …
Optimised annealed sequential monte carlo samplers
Annealed Sequential Monte Carlo (SMC) samplers are special cases of SMC samplers
where the sequence of distributions can be embedded in a smooth path of distributions …
where the sequence of distributions can be embedded in a smooth path of distributions …
Provable Convergence and Limitations of Geometric Tempering for Langevin Dynamics
Geometric tempering is a popular approach to sampling from challenging multi-modal
probability distributions by instead sampling from a sequence of distributions which …
probability distributions by instead sampling from a sequence of distributions which …
Theoretical Guarantees for Variational Inference with Fixed-Variance Mixture of Gaussians
Variational inference (VI) is a popular approach in Bayesian inference, that looks for the best
approximation of the posterior distribution within a parametric family, minimizing a loss that …
approximation of the posterior distribution within a parametric family, minimizing a loss that …
Stochastic mirror descent for nonparametric adaptive importance sampling
P Bianchi, B Delyon, V Priser, F Portier - arXiv preprint arXiv:2409.13272, 2024 - arxiv.org
This paper addresses the problem of approximating an unknown probability distribution with
density $ f $--which can only be evaluated up to an unknown scaling factor--with the help of …
density $ f $--which can only be evaluated up to an unknown scaling factor--with the help of …
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 …
describing evolutionary processes in biology and sequential Bayesian learning methods …
Inclusive KL Minimization: A Wasserstein-Fisher-Rao Gradient Flow Perspective
JJ Zhu - arXiv preprint arXiv:2411.00214, 2024 - arxiv.org
Otto's (2001) Wasserstein gradient flow of the exclusive KL divergence functional provides a
powerful and mathematically principled perspective for analyzing learning and inference …
powerful and mathematically principled perspective for analyzing learning and inference …
Ensemble Kalman inversion approximate Bayesian computation
RG Everitt - arXiv preprint arXiv:2407.18721, 2024 - arxiv.org
Approximate Bayesian computation (ABC) is the most popular approach to inferring
parameters in the case where the data model is specified in the form of a simulator. It is not …
parameters in the case where the data model is specified in the form of a simulator. It is not …