Removing systematic errors for exoplanet search via latent causes
Machine Learning
2015-05-13 v1 Earth and Planetary Astrophysics
Instrumentation and Methods for Astrophysics
Machine Learning
Abstract
We describe a method for removing the effect of confounders in order to reconstruct a latent quantity of interest. The method, referred to as half-sibling regression, is inspired by recent work in causal inference using additive noise models. We provide a theoretical justification and illustrate the potential of the method in a challenging astronomy application.
Keywords
Cite
@article{arxiv.1505.03036,
title = {Removing systematic errors for exoplanet search via latent causes},
author = {Bernhard Schölkopf and David W. Hogg and Dun Wang and Daniel Foreman-Mackey and Dominik Janzing and Carl-Johann Simon-Gabriel and Jonas Peters},
journal= {arXiv preprint arXiv:1505.03036},
year = {2015}
}
Comments
Extended version of a paper appearing in the Proceedings of the 32nd International Conference on Machine Learning, Lille, France, 2015