MultiMatch: Multi-task Learning for Semi-supervised Domain Generalization

arXiv:2208.05853v2 Announce Type: replace
Abstract: Domain generalization (DG) aims at learning a model on source domains to well generalize on the unseen target domain. Although it has achieved great success, most of existing methods require the label information for all training samples in source domains, which is time-consuming and expensive in the real-world application. In this paper, we resort to solving the semi-supervised domain generalization (SSDG) task, where there are a few label information in each source domain. To address …

abstract arxiv cs.cv domain domains information multi-task learning samples semi-supervised success training type world