Deterministic Langevin Monte Carlo with Normalizing Flows for Bayesian Inference

Abstract

We propose a general purpose Bayesian inference algorithm for expensive likelihoods, replacing the stochastic term in the Langevin equation with a deterministic density gradient term. The particle density is evaluated from the current particle positions using a Normalizing Flow (NF), which is differentiable and has good generalization properties in high dimensions. We take advantage of NF preconditioning and NF based Metropolis-Hastings updates for a faster convergence. We show on various examples that the method is competitive against state of the art sampling methods.

PDF NeurIPS OpenReview Semantic Scholar

Cite

Text

Grumitt et al. "Deterministic Langevin Monte Carlo with Normalizing Flows for Bayesian Inference." Neural Information Processing Systems, 2022.

Markdown

[Grumitt et al. "Deterministic Langevin Monte Carlo with Normalizing Flows for Bayesian Inference." Neural Information Processing Systems, 2022.](https://mlanthology.org/neurips/2022/grumitt2022neurips-deterministic/)

BibTeX

@inproceedings{grumitt2022neurips-deterministic,
  title     = {{Deterministic Langevin Monte Carlo with Normalizing Flows for Bayesian Inference}},
  author    = {Grumitt, Richard and Dai, Biwei and Seljak, Uros},
  booktitle = {Neural Information Processing Systems},
  year      = {2022},
  url       = {https://mlanthology.org/neurips/2022/grumitt2022neurips-deterministic/}
}