On a Modification to the Mean Field EM Algorithm in Factorial Learning

Abstract

A modification is described to the use of mean field approxima(cid:173) tions in the E step of EM algorithms for analysing data from latent structure models, as described by Ghahramani (1995), among oth(cid:173) ers. The modification involves second-order Taylor approximations to expectations computed in the E step. The potential benefits of the method are illustrated using very simple latent profile models.

Cite

Text

Dunmur and Titterington. "On a Modification to the Mean Field EM Algorithm in Factorial Learning." Neural Information Processing Systems, 1996.

Markdown

[Dunmur and Titterington. "On a Modification to the Mean Field EM Algorithm in Factorial Learning." Neural Information Processing Systems, 1996.](https://mlanthology.org/neurips/1996/dunmur1996neurips-modification/)

BibTeX

@inproceedings{dunmur1996neurips-modification,
  title     = {{On a Modification to the Mean Field EM Algorithm in Factorial Learning}},
  author    = {Dunmur, A. P. and Titterington, D. M.},
  booktitle = {Neural Information Processing Systems},
  year      = {1996},
  pages     = {431-437},
  url       = {https://mlanthology.org/neurips/1996/dunmur1996neurips-modification/}
}