Hierarchical IFA Belief Networks

Hagai Attias
Proceedings of the Seventh International Workshop on Artificial Intelligence and Statistics, PMLR R2:1-9, 1999.

Abstract

We introduce a new real-valued belief network, which is a multilayer generalization of independent factor analysis (IFA). At each level, this network extracts real-valued latent variables that are non-linear functions of the input data with a highly adaptive functional form, resulting in a hierarchical distributed representation of these data. The network is based on a probabilistic generative model, constructed by cascading single-layer IFA models. Whereas exact maximum-likelihood learning for this model is intractable, we present and demonstrate an algorithm that maximizes a lower bound on the likelihood. This algorithm is developed by formulating a variational approach to hierarchical IFA networks.

Cite this Paper

BibTeX
@InProceedings{pmlr-vR2-attias99a, title = {Hierarchical {IFA} Belief Networks}, author = {Attias, Hagai}, booktitle = {Proceedings of the Seventh International Workshop on Artificial Intelligence and Statistics}, pages = {1--9}, year = {1999}, editor = {Heckerman, David and Whittaker, Joe}, volume = {R2}, series = {Proceedings of Machine Learning Research}, month = {03--06 Jan}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/r2/attias99a/attias99a.pdf}, url = {https://proceedings.mlr.press/r2/attias99a.html}, abstract = {We introduce a new real-valued belief network, which is a multilayer generalization of independent factor analysis (IFA). At each level, this network extracts real-valued latent variables that are non-linear functions of the input data with a highly adaptive functional form, resulting in a hierarchical distributed representation of these data. The network is based on a probabilistic generative model, constructed by cascading single-layer IFA models. Whereas exact maximum-likelihood learning for this model is intractable, we present and demonstrate an algorithm that maximizes a lower bound on the likelihood. This algorithm is developed by formulating a variational approach to hierarchical IFA networks.}, note = {Reissued by PMLR on 20 August 2020.} }
Endnote
%0 Conference Paper %T Hierarchical IFA Belief Networks %A Hagai Attias %B Proceedings of the Seventh International Workshop on Artificial Intelligence and Statistics %C Proceedings of Machine Learning Research %D 1999 %E David Heckerman %E Joe Whittaker %F pmlr-vR2-attias99a %I PMLR %P 1--9 %U https://proceedings.mlr.press/r2/attias99a.html %V R2 %X We introduce a new real-valued belief network, which is a multilayer generalization of independent factor analysis (IFA). At each level, this network extracts real-valued latent variables that are non-linear functions of the input data with a highly adaptive functional form, resulting in a hierarchical distributed representation of these data. The network is based on a probabilistic generative model, constructed by cascading single-layer IFA models. Whereas exact maximum-likelihood learning for this model is intractable, we present and demonstrate an algorithm that maximizes a lower bound on the likelihood. This algorithm is developed by formulating a variational approach to hierarchical IFA networks. %Z Reissued by PMLR on 20 August 2020.
APA
Attias, H.. (1999). Hierarchical IFA Belief Networks. Proceedings of the Seventh International Workshop on Artificial Intelligence and Statistics, in Proceedings of Machine Learning Research R2:1-9 Available from https://proceedings.mlr.press/r2/attias99a.html. Reissued by PMLR on 20 August 2020.

Related Material