Factorizing Multivariate Function Classes
Abstract
The mathematical framework for factorizing equivalence classes of multivariate functions is formulated in this paper. Independent component analysis is shown to be a special case of this decompo(cid:173) sition. Using only the local geometric structure of a class repre(cid:173) sentative, we derive an analytic solution for the factorization. We demonstrate the factorization solution with numerical experiments and present a preliminary tie to decorrelation.
Cite
Text
Lin. "Factorizing Multivariate Function Classes." Neural Information Processing Systems, 1997.Markdown
[Lin. "Factorizing Multivariate Function Classes." Neural Information Processing Systems, 1997.](https://mlanthology.org/neurips/1997/lin1997neurips-factorizing/)BibTeX
@inproceedings{lin1997neurips-factorizing,
title = {{Factorizing Multivariate Function Classes}},
author = {Lin, Juan K.},
booktitle = {Neural Information Processing Systems},
year = {1997},
pages = {563-569},
url = {https://mlanthology.org/neurips/1997/lin1997neurips-factorizing/}
}