Stacking for machine learning redshifts applied to SDSS galaxies

no code implementations • 19 Feb 2016 • Roman Zitlau, Ben Hoyle, Kerstin Paech, Jochen Weller, Markus Michael Rau, Stella Seitz

We observe a significant improvement of between 1. 9% and 21% on all computed metrics when stacking is applied to weak learners (such as SOMs and decision trees).

BIG-bench Machine Learning

Anomaly detection for machine learning redshifts applied to SDSS galaxies

no code implementations • 27 Mar 2015 • Ben Hoyle, Markus Michael Rau, Kerstin Paech, Christopher Bonnett, Stella Seitz, Jochen Weller

We present an analysis of anomaly detection for machine learning redshift estimation.

Cosmology and Nongalactic Astrophysics

Feature importance for machine learning redshifts applied to SDSS galaxies

no code implementations • 17 Oct 2014 • Ben Hoyle, Markus Michael Rau, Roman Zitlau, Stella Seitz, Jochen Weller

When compared to the SDSS photometric redshifts, the RDF machine learning redshifts both decreases the standard deviation of residuals scaled by 1/(1+z) by 36% from 0. 066 to 0. 041, and decreases the fraction of catastrophic outliers by 57% from 2. 32% to 0. 99%.

Instrumentation and Methods for Astrophysics Cosmology and Nongalactic Astrophysics