Department of Mathematical Sciences
|Thursday, October 21, 2021
|1:15pm – 2:15pm
|Baozhen Wang, Binghamton University
|Covariate Shift by Kernel Mean Matching
Given sets of observations of training and test data, the authors consider the problem of re-weighting the training data such that its distribution more closely matches that of the test data. They achieve this goal by matching covariate distributions between training and test sets in a high dimensional feature space (specifically, a reproducing kernel Hilbert space). This approach does not require distribution estimation. Instead, the sample weights are obtained by a simple quadratic programming procedure.