A stochastic version of Stein Variational Gradient Descent for efficient sampling
arXiv:1902.03394 · doi:10.2140/camcos.2020.15.37
Abstract
We propose in this work RBM-SVGD, a stochastic version of Stein Variational Gradient Descent (SVGD) method for efficiently sampling from a given probability measure and thus useful for Bayesian inference. The method is to apply the Random Batch Method (RBM) for interacting particle systems proposed by Jin et al to the interacting particle systems in SVGD. While keeping the behaviors of SVGD, it reduces the computational cost, especially when the interacting kernel has long range. Numerical examples verify the efficiency of this new version of SVGD.
References in corpus (10)
- Variational Inference: A Review for Statisticians
- Variational Inference with Normalizing Flows
- Stein Variational Gradient Descent: A General Purpose Bayesian Inference Algorithm
- Data Assimilation: A Mathematical Introduction
- AdaGrad stepsizes: Sharp convergence over nonconvex landscapes
- On the Random Batch Method for second order interacting particle systems
- Stein Variational Gradient Descent as Gradient Flow
- Provable Bayesian Inference via Particle Mirror Descent
- A Unified Particle-Optimization Framework for Scalable Bayesian Sampling
- A Stein variational Newton method
Cited by in corpus (17)
- On the Random Batch Method for second order interacting particle systems
- On the geometry of Stein variational gradient descent
- Random batch particle methods for the homogeneous Landau equation
- Stein variational gradient descent with local approximations
- Particle-based Energetic Variational Inference
- Stochastic Stein Discrepancies
- A random batch Ewald method for particle systems with Coulomb interactions
- Convergence of Random Batch Method for interacting particles with disparate species and weights
- Learning interaction kernels in stochastic systems of interacting particles from multiple trajectories
- Convergence of Stein Variational Gradient Descent under a Weaker Smoothness Condition
- Random-batch method for multi-species stochastic interacting particle systems
- Random Batch Algorithms for Quantum Monte Carlo simulations
- Some random batch particle methods for the Poisson-Nernst-Planck and Poisson-Boltzmann equations
- A splitting Hamiltonian Monte Carlo method for efficient sampling
- Stein variational reduced basis Bayesian inversion
- On the mean field limit of the Random Batch Method for interacting particle systems
- Random-batch list algorithm for short-range molecular dynamics simulations