Learning Coordinate Covariances via Gradients

Abstract

We introduce an algorithm that learns gradients from samples in the supervised learning framework. An error analysis is given for the convergence of the gradient estimated by the algorithm to the true gradient. The utility of the algorithm for the problem of variable selection as well as determining variable covariance is illustrated on simulated data as well as two gene expression data sets. For square loss we provide a very efficient implementation with respect to both memory and time.

Cite

Text

Mukherjee and Zhou. "Learning Coordinate Covariances via Gradients." Journal of Machine Learning Research, 2006.

Markdown

[Mukherjee and Zhou. "Learning Coordinate Covariances via Gradients." Journal of Machine Learning Research, 2006.](https://mlanthology.org/jmlr/2006/mukherjee2006jmlr-learning/)

BibTeX

@article{mukherjee2006jmlr-learning,
  title     = {{Learning Coordinate Covariances via Gradients}},
  author    = {Mukherjee, Sayan and Zhou, Ding-Xuan},
  journal   = {Journal of Machine Learning Research},
  year      = {2006},
  pages     = {519-549},
  volume    = {7},
  url       = {https://mlanthology.org/jmlr/2006/mukherjee2006jmlr-learning/}
}