A new similarity measure for covariate shift with applications to nonparametric regression

Reese Pathak · Cong Ma · Martin Wainwright

Hall E #1114

Keywords: [ T: Domain Adaptation and Transfer Learning ]

[ Abstract ]
[ Poster [ Paper PDF
Wed 20 Jul 3:30 p.m. PDT — 5:30 p.m. PDT
Oral presentation: T: Learning Theory/Domain Adaptation
Wed 20 Jul 7:30 a.m. PDT — 9 a.m. PDT


We study covariate shift in the context of nonparametric regression. We introduce a new measure of distribution mismatch between the source and target distributions using the integrated ratio of probabilities of balls at a given radius. We use the scaling of this measure with respect to the radius to characterize the minimax rate of estimation over a family of Hölder continuous functions under covariate shift. In comparison to the recently proposed notion of transfer exponent, this measure leads to a sharper rate of convergence and is more fine-grained. We accompany our theory with concrete instances of covariate shift that illustrate this sharp difference.

Chat is not available.