Dataset Shift in Machine Learning Book

I Introduction to dataset shift 1

1 When training and test sets are different: characterizing learning transfer Amos Storkey Storkey, Amos 3

2 Projection and projectability David Corfield Corfield, David 29

II Theoretical views on dataset and covariate shift 39

3 Binary classification under sample selection bias Matthias Hein Hein, Matthias 41

4 On Bayesian transduction: implications for the covariate shift problem Lars Kai Hansen Hansen, Lars Kai 65

5 On the training/test distributions gap: a data representation learning framework Shai Ben-David Ben-David, Shai 73

III Algorithms for covariate shift 85

6 Geometry of covariate shift with applications to active learning Takafumi Kanamori Kanamori, Takafumi Hidetoshi Shimodaira Shimodaira, Hidetoshi 87

7 A conditional expectation approach to model selection and active learning under covariate shift Masashi Sugiyama Sugiyama, Masashi Neil Rubens Rubens, Neil Klaus-Robert Muller Muller, Klaus-Robert 107

8 Covariate shift by kernel mean matching Arthur Grellon Grellon, Arthur Alex Smola Smola, Alex Jiayuan Huang Huang, Jiayuan Marcel Schmittfull Schmittfull, Marcel Karsten Borgwardt Borgwardt, Karsten Bernhard Scholkopf Scholkopf, Bernhard 131

9 Discriminative learning under covariate shift with a single optimization problem Steffen Bickel Bickel, Steffen Michael Bruckner Bruckner, Michael Tobias Scheffer Scheffer, Tobias 161

10 An adversarial view of covariate shift and a minimax approach Amir Globerson Globerson, Amir Choon Hui Teo Teo, Choon Hui Alex Smola Smola, Alex Sam Roweis Roweis, Sam 179

IV Discussion 199

11 Author comments Hidetoshi Shimodaira Shimodaira, Hidetoshi Masashi Sugiyama Sugiyama,Masashi Amos Storkey Storkey, Amos Arthur Gretton Gretton, Arthur Shai-Ben David David, Shai-Ben 201

References 207

Notation and symbols 219

Contributors 223

Index 227