Stratified Learning: A General-Purpose Statistical Method for Improved Learning under Covariate Shift
arXiv:2106.11211 · doi:10.1002/sam.11643
Abstract
We propose a simple, statistically principled, and theoretically justified method to improve supervised learning when the training set is not representative, a situation known as covariate shift. We build upon a well-established methodology in causal inference, and show that the effects of covariate shift can be reduced or eliminated by conditioning on propensity scores. In practice, this is achieved by fitting learners within strata constructed by partitioning the data based on the estimated propensity scores, leading to approximately balanced covariates and much-improved target prediction. We demonstrate the effectiveness of our general-purpose method on two contemporary research questions in cosmology, outperforming state-of-the-art importance weighting methods. We obtain the best reported AUC (0.958) on the updated "Supernovae photometric classification challenge", and we improve upon existing conditional density estimation of galaxy redshift from Sloan Data Sky Survey (SDSS) data.
References in corpus (12)
- Observational Evidence from Supernovae for an Accelerating Universe and a Cosmological Constant
- Measurements of Omega and Lambda from 42 High-Redshift Supernovae
- A review of domain adaptation without target labels
- Estimating the Redshift Distribution of Faint Galaxy Samples
- Photometric Supernova Classification With Machine Learning
- Results from the Supernova Photometric Classification Challenge
- Models and Simulations for the Photometric LSST Astronomical Time Series Classification Challenge (PLAsTiCC)
- Avocado: Photometric Classification of Astronomical Transients with Gaussian Process Augmentation
- Photometric Redshift Probability Distributions for Galaxies in the SDSS DR8
- PELICAN: deeP architecturE for the LIght Curve ANalysis
- Semi-supervised Learning for Photometric Supernova Classification
- STACCATO: A Novel Solution to Supernova Photometric Classification with Biased Training Sets