Search Machine Learning Repository: Domain Adaptation under Target and Conditional Shift
Authors: Kun Zhang, Bernhard Schlkopf, Krikamol Muandet and Zhikun Wang
Conference: Proceedings of the 30th International Conference on Machine Learning (ICML-13)
Year: 2013
Pages: 819-827
Abstract: Let $X$ denote the feature and $Y$ the target. We consider domain adaptation under three possible scenarios: (1) the marginal $P_Y$ changes, while the conditional $P_{X|Y}$ stays the same ({\it target shift}), (2) the marginal $P_Y$ is fixed, while the conditional $P_{X|Y}$ changes with certain constraints ({\it conditional shift}), and (3) the marginal $P_{Y}$ changes, and the conditional $P_{X|Y}$ changes with constraints ({\it generalized target shift}). Using background knowledge, causal interpretations allow us to determine the correct situation for a problem at hand. We exploit importance reweighting or sample transformation to find the learning machine that works well on test data, and propose to estimate the weights or transformations by {\it reweighting or transforming training data to reproduce the covariate distribution} on the test domain. Thanks to kernel embedding of conditional as well as marginal distributions, the proposed approaches avoid distribution estimation, and are applicable for high-dimensional problems. Numerical evaluations on synthetic and real-world datasets demonstrate the effectiveness of the proposed framework.
[pdf] [BibTeX]

authors venues years
Suggest Changes to this paper.
Brought to you by the WUSTL Machine Learning Group. We have open faculty positions (tenured and tenure-track).