TY - JOUR
T1 - Feature-level domain adaptation
AU - Kouw, Wouter M.
AU - Van Der Maaten, Laurens J P
AU - Krijthe, Jesse H.
AU - Loog, Marco
PY - 2016
Y1 - 2016
N2 - Domain adaptation is the supervised learning setting in which the training and test data are sampled from different distributions: training data is sampled from a source domain, whilst test data is sampled from a target domain. This paper proposes and studies an approach, called feature-level domain adaptation (flda), that models the dependence between the two domains by means of a feature-level transfer model that is trained to describe the transfer from source to target domain. Subsequently, we train a domain-adapted classifier by minimizing the expected loss under the resulting transfer model. For linear classifiers and a large family of loss functions and transfer models, this expected loss can be computed or approximated analytically, and minimized efficiently. Our empirical evaluation of flda focuses on problems comprising binary and count data in which the transfer can be naturally modeled via a dropout distribution, which allows the classiffier to adapt to differences in the marginal probability of features in the source and the target domain. Our experiments on several real-world problems show that flda performs on par with state-of-the-art domainadaptation techniques.
AB - Domain adaptation is the supervised learning setting in which the training and test data are sampled from different distributions: training data is sampled from a source domain, whilst test data is sampled from a target domain. This paper proposes and studies an approach, called feature-level domain adaptation (flda), that models the dependence between the two domains by means of a feature-level transfer model that is trained to describe the transfer from source to target domain. Subsequently, we train a domain-adapted classifier by minimizing the expected loss under the resulting transfer model. For linear classifiers and a large family of loss functions and transfer models, this expected loss can be computed or approximated analytically, and minimized efficiently. Our empirical evaluation of flda focuses on problems comprising binary and count data in which the transfer can be naturally modeled via a dropout distribution, which allows the classiffier to adapt to differences in the marginal probability of features in the source and the target domain. Our experiments on several real-world problems show that flda performs on par with state-of-the-art domainadaptation techniques.
KW - Covariate shift
KW - Domain adaptation
KW - Risk minimization
KW - Transfer learning
UR - http://www.scopus.com/inward/record.url?scp=84995487610&partnerID=8YFLogxK
M3 - Journal article
AN - SCOPUS:84995487610
SN - 1533-7928
VL - 17
JO - Journal of Machine Learning Research
JF - Journal of Machine Learning Research
M1 - 171
ER -