Correcting Sample Selection Bias by Unlabeled Data

Citations Over TimeTop 1% of 2007 papers

Abstract

We consider the scenario where training and test data are drawn from different distributions, commonly referred to as sample selection bias.Most algorithms for this setting try to first recover sampling distributions and then make appropriate corrections based on the distribution estimate.We present a nonparametric method which directly produces resampling weights without distribution estimation.Our method works by matching distributions between training and testing sets in feature space.Experimental results demonstrate that our method works well in practice.

Related Papers

• → Observed characteristics and teacher quality: Impacts of sample selection on a value added model(2011)30 cited
• → Estimates of the Average Strength of Natural Selection Are Not Inflated by Sampling Error or Publication Bias(2007)22 cited
• → Selection Bias(2009)
• → Estimates of the Average Strength of Natural Selection Are Not Inflated by Sampling Error or Publication Bias(2007)