Learning in high dimensions: Modular Mixture Models

Hagai Attias
Proceedings of the Eighth International Workshop on Artificial Intelligence and Statistics, PMLR R3:8-12, 2001.

Abstract

We present a new approach to learning prob- abilistic models for high dimensional data. This approach divides the data dimensions into low dimensional subspaces, and learns a separate mixture model for each subspace. The models combine in a principled manner to form a flexible modular network that pro- duces a total density estimate. We derive and demonstrate an iterative learning algorithm that uses only local information.

Cite this Paper

BibTeX
@InProceedings{pmlr-vR3-attias01a, title = {Learning in high dimensions: Modular Mixture Models}, author = {Attias, Hagai}, booktitle = {Proceedings of the Eighth International Workshop on Artificial Intelligence and Statistics}, pages = {8--12}, year = {2001}, editor = {Richardson, Thomas S. and Jaakkola, Tommi S.}, volume = {R3}, series = {Proceedings of Machine Learning Research}, month = {04--07 Jan}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/r3/attias01a/attias01a.pdf}, url = {https://proceedings.mlr.press/r3/attias01a.html}, abstract = {We present a new approach to learning prob- abilistic models for high dimensional data. This approach divides the data dimensions into low dimensional subspaces, and learns a separate mixture model for each subspace. The models combine in a principled manner to form a flexible modular network that pro- duces a total density estimate. We derive and demonstrate an iterative learning algorithm that uses only local information.}, note = {Reissued by PMLR on 31 March 2021.} }
Endnote
%0 Conference Paper %T Learning in high dimensions: Modular Mixture Models %A Hagai Attias %B Proceedings of the Eighth International Workshop on Artificial Intelligence and Statistics %C Proceedings of Machine Learning Research %D 2001 %E Thomas S. Richardson %E Tommi S. Jaakkola %F pmlr-vR3-attias01a %I PMLR %P 8--12 %U https://proceedings.mlr.press/r3/attias01a.html %V R3 %X We present a new approach to learning prob- abilistic models for high dimensional data. This approach divides the data dimensions into low dimensional subspaces, and learns a separate mixture model for each subspace. The models combine in a principled manner to form a flexible modular network that pro- duces a total density estimate. We derive and demonstrate an iterative learning algorithm that uses only local information. %Z Reissued by PMLR on 31 March 2021.
APA
Attias, H.. (2001). Learning in high dimensions: Modular Mixture Models. Proceedings of the Eighth International Workshop on Artificial Intelligence and Statistics, in Proceedings of Machine Learning Research R3:8-12 Available from https://proceedings.mlr.press/r3/attias01a.html. Reissued by PMLR on 31 March 2021.

Related Material