AISTATS 2023poster33 citations
Particle algorithms for maximum likelihood training of latent variable models
Juan Kuntz, Jen Ning Lim, Adam M. Johansen
Abstract
Neal and Hinton (1998) recast maximum likelihood estimation of any given latent variable model as the minimization of a free energy functional F, and the EM algorithm as coordinate descent applied to F. Here, we explore alternative ways to optimize the functional. In particular, we identify various gradient flows associated with F and show that their limits coincide with F’s stationary points. By discretizing the flows, we obtain practical particle-based algorithms for maximum likelihood estimation in broad classes of latent variable models. The novel algorithms scale to high-dimensional settings and perform well in numerical experiments.
BibTeX
@InProceedings{pmlr-v206-kuntz23a,
title = {Particle algorithms for maximum likelihood training of latent variable models},
author = {Kuntz, Juan and Lim, Jen Ning and Johansen, Adam M.},
booktitle = {Proceedings of The 26th International Conference on Artificial Intelligence and Statistics},
pages = {5134--5180},
year = {2023},
editor = {Ruiz, Francisco and Dy, Jennifer and van de Meent, Jan-Willem},
volume = {206},
series = {Proceedings of Machine Learning Research},
month = {25--27 Apr},
publisher = {PMLR},
pdf = {https://proceedings.mlr.press/v206/kuntz23a/kuntz23a.pdf},
url = {https://proceedings.mlr.press/v206/kuntz23a.html},
abstract = {Neal and Hinton (1998) recast maximum likelihood estimation of any given latent variable model as the minimization of a free energy functional F, and the EM algorithm as coordinate descent applied to F. Here, we explore alternative ways to optimize the functional. In particular, we identify various gradient flows associated with F and show that their limits coincide with F’s stationary points. By discretizing the flows, we obtain practical particle-based algorithms for maximum likelihood estimation in broad classes of latent variable models. The novel algorithms scale to high-dimensional settings and perform well in numerical experiments.}
}