English

Moment-Based Variational Inference for Stochastic Differential Equations

Machine Learning 2021-03-02 v1 Machine Learning

Abstract

Existing deterministic variational inference approaches for diffusion processes use simple proposals and target the marginal density of the posterior. We construct the variational process as a controlled version of the prior process and approximate the posterior by a set of moment functions. In combination with moment closure, the smoothing problem is reduced to a deterministic optimal control problem. Exploiting the path-wise Fisher information, we propose an optimization procedure that corresponds to a natural gradient descent in the variational parameters. Our approach allows for richer variational approximations that extend to state-dependent diffusion terms. The classical Gaussian process approximation is recovered as a special case.

Keywords

Cite

@article{arxiv.2103.00988,
  title  = {Moment-Based Variational Inference for Stochastic Differential Equations},
  author = {Christian Wildner and Heinz Koeppl},
  journal= {arXiv preprint arXiv:2103.00988},
  year   = {2021}
}

Comments

Appearing in Proceedings of the 24th International Conference on Artificial Intelligence and Statistics (AISTATS) 2021, San Diego, California, USA. PMLR: Volume 130

R2 v1 2026-06-23T23:37:00.379Z