Correcting Sample Selection Bias by Unlabeled Data

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.

[1]  W. Hoeffding.  Probability Inequalities for sums of Bounded Random Variables , 1963 .

[2]  J. Heckman.  Sample selection bias as a specification error , 1979 .

[3]  W. Greene.  Sample Selection Bias as a Specification Error: Comment , 1981 .

[4]  R. M. Dudley,et al.  Real Analysis and Probability , 1989 .

[5]  Colin McDiarmid,et al.  Surveys in Combinatorics, 1989: On the method of bounded differences , 1989 .

[6]  N. Fisher,et al.  Probability Inequalities for Sums of Bounded Random Variables , 1994 .

[7]  Herbert Gish,et al.  Speaker identification via support vector classifiers , 1996, 1996 IEEE International Conference on Acoustics, Speech, and Signal Processing Conference Proceedings.

[8]  David Barber,et al.  Bayesian Classification With Gaussian Processes , 1998, IEEE Trans. Pattern Anal. Mach. Intell..

[9]  H. Shimodaira,et al.  Improving predictive inference under covariate shift by weighting the log-likelihood function , 2000 .

[10]  Bernhard Schölkopf,et al.  Estimating the Support of a High-Dimensional Distribution , 2001, Neural Computation.

[11]  Ingo Steinwart,et al.  On the Influence of the Kernel on the Consistency of Support Vector Machines , 2002, J. Mach. Learn. Res..

[12]  Carsten O. Peterson,et al.  Estrogen receptor status in breast cancer is associated with remarkably distinct gene expression patterns. , 2001, Cancer research.

[13]  R. Spang,et al.  Predicting the clinical status of human breast cancer by using gene expression profiles , 2001, Proceedings of the National Academy of Sciences of the United States of America.

[14]  Ingo Steinwart,et al.  Support Vector Machines are Universally Consistent , 2002, J. Complex..

[15]  Shahar Mendelson,et al.  A Few Notes on Statistical Learning Theory , 2002, Machine Learning Summer School.

[16]  E. Lander,et al.  Gene expression correlates of clinical prostate cancer behavior. , 2002, Cancer cell.

[17]  Ji Zhu,et al.  A Method for Inferring Label Sampling Mechanisms in Semi-Supervised Learning , 2004, NIPS.

[18]  Bianca Zadrozny,et al.  Learning and evaluating classifiers under sample selection bias , 2004, ICML.

[19]  Yi Lin,et al.  Support Vector Machines for Classification in Nonstandard Situations , 2002, Machine Learning.

[20]  Masashi Sugiyama,et al.  Input-dependent estimation of generalization error under covariate shift , 2005 .

[21]  K. Müller,et al.  Generalization Error Estimation under Covariate Shift , 2005 .

[22]  Roland Eils,et al.  Cross-platform analysis of cancer microarray data improves gene expression based classification of phenotypes , 2005, BMC Bioinformatics.

[23]  Miroslav Dudík,et al.  Correcting sample selection bias in maximum entropy density estimation , 2005, NIPS.

[24]  Thomas Hofmann,et al.  Large Margin Methods for Structured and Interdependent Output Variables , 2005, J. Mach. Learn. Res..

[25]  Bernhard Schölkopf,et al.  A Kernel Method for the Two-Sample-Problem , 2006, NIPS.

[26]  S. Zacks,et al.  Journal of Statistical Planning and Inference , 2016 .