Discrete gradients for computational Bayesian inference
From MaRDI portal
Publication:2297877
Abstract: In this paper, we exploit the gradient flow structure of continuous-time formulations of Bayesian inference in terms of their numerical time-stepping. We focus on two particular examples, namely, the continuous-time ensemble Kalman-Bucy filter and a particle discretisation of the Fokker-Planck equation associated to Brownian dynamics. Both formulations can lead to stiff differential equations which require special numerical methods for their efficient numerical implementation. We compare discrete gradient methods to alternative semi-implicit and other iterative implementations of the underlying Bayesian inference problems.
Recommendations
- A note on parametric Bayesian inference via gradient flows
- Fokker-Planck particle systems for Bayesian inference: computational approaches
- Particle Metropolis-Hastings using gradient and Hessian information
- Ensemble Kalman filter based sequential Monte Carlo sampler for sequential Bayesian inference
- Signal processing problems on function space: Bayesian formulation, stochastic PDEs and effective MCMC methods
Cites work
- A Deterministic Approximation of Diffusion Equations Using Particles
- A dynamical systems framework for intermittent data assimilation
- A hybrid ensemble transform particle filter for nonlinear and spatially extended dynamical systems
- A new class of particle methods
- A strongly convergent numerical scheme from ensemble Kalman inversion
- Approximate McKean-Vlasov representations for a class of SPDEs
- Data Assimilation
- Data assimilation: the Schrödinger perspective
- Deterministic diffusion of particles
- Deterministic Nonperiodic Flow
- Enhancing energy conserving methods
- Ensemble Kalman inversion: a derivative-free technique for machine learning tasks
- Geometric integration using discrete gradients
- scientific article; zbMATH DE number 1631958 (Why is no real title available?)
- scientific article; zbMATH DE number 671796 (Why is no real title available?)
- scientific article; zbMATH DE number 5060482 (Why is no real title available?)
- Interacting Langevin diffusions: gradient structure and ensemble Kalman sampler
- Inverse Problem Theory and Methods for Model Parameter Estimation
- Levenberg-Marquardt forms of the iterative ensemble smoother for efficient history matching and uncertainty quantification
- Long-time stability and accuracy of the ensemble Kalman-Bucy filter for fully observed processes and small measurement noise
- Numerical Methods for Evolutionary Differential Equations
- Online natural gradient as a Kalman filter
- Statistical and computational inverse problems.
- Stochastic processes and applications. Diffusion processes, the Fokker-Planck and Langevin equations
- The ensemble Kalman filter for combined state and parameter estimation
- Time integration and discrete Hamiltonian systems
Cited in
(18)- A geometric integration approach to nonsmooth, nonconvex optimisation
- A note on parametric Bayesian inference via gradient flows
- The Bayesian update: variational formulations and gradient flows
- Preface. Special issue in honor of Reinout Quispel
- scientific article; zbMATH DE number 6860839 (Why is no real title available?)
- Affine invariant interacting Langevin dynamics for Bayesian inference
- Fokker-Planck particle systems for Bayesian inference: computational approaches
- Neural parametric Fokker-Planck equation
- Ensemble inference methods for models with noisy and expensive likelihoods
- Interacting Langevin diffusions: gradient structure and ensemble Kalman sampler
- Data assimilation: the Schrödinger perspective
- Projected Wasserstein Gradient Descent for High-Dimensional Bayesian Inference
- A Continuation Method in Bayesian Inference
- Sequential discretization schemes for a class of stochastic differential equations and their application to Bayesian filtering
- On the ensemble Kalman inversion under inequality constraints
- Preserving nonlinear constraints in variational flow filtering data assimilation
- Ensemble variational Fokker-Planck methods for data assimilation
- Gradient-free sequential Bayesian experimental design via interacting particle systems
This page was built for publication: Discrete gradients for computational Bayesian inference
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2297877)
