Distance Covariance Analysis
Abstract
We propose a dimensionality reduction method to identify linear projections that capture interactions between two or more sets of variables. The method, distance covariance analysis (DCA), can detect both linear and nonlinear relationships, and can take dependent variables into account. On previous testbeds and a new testbed that systematically assesses the ability to detect both linear and nonlinear interactions, DCA performs better than or comparable to existing methods, while being one of the fastest methods. To showcase the versatility of DCA, we also applied it to three different neurophysiological datasets.
Cite
Text
Cowley et al. "Distance Covariance Analysis." International Conference on Artificial Intelligence and Statistics, 2017.Markdown
[Cowley et al. "Distance Covariance Analysis." International Conference on Artificial Intelligence and Statistics, 2017.](https://mlanthology.org/aistats/2017/cowley2017aistats-distance/)BibTeX
@inproceedings{cowley2017aistats-distance,
title = {{Distance Covariance Analysis}},
author = {Cowley, Benjamin and Semedo, João D. and Zandvakili, Amin and Smith, Matthew A. and Kohn, Adam and Yu, Byron M.},
booktitle = {International Conference on Artificial Intelligence and Statistics},
year = {2017},
pages = {242-251},
url = {https://mlanthology.org/aistats/2017/cowley2017aistats-distance/}
}