Back to Search
Start Over
Removing systematic errors for exoplanet search via latent causes
- Publication Year :
- 2015
-
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.<br />Comment: Extended version of a paper appearing in the Proceedings of the 32nd International Conference on Machine Learning, Lille, France, 2015
Details
- Database :
- arXiv
- Publication Type :
- Report
- Accession number :
- edsarx.1505.03036
- Document Type :
- Working Paper