1 paper
Seungcheol Park, Huiwen Xu, Taehun Kim +3
Given a set of heterogeneous source datasets with their classifiers, how can we quickly find the most useful source dataset for a specific target task? We address the problem of me…