2014/11/17 by Basura Fernando, Fernando, Basura, Tatiana Tommasi +3
Biochemistry, Genetics and Molecular Biology · Computer Science · Medicine · #Cancer-related molecular mechanisms research #Computer Vision and Pattern Recognition (cs.CV) #Domain Adaptation and Few-Shot Learning #FOS: Computer and information sciences #Machine Learning (cs.LG) #Respiratory viral infections research #cs.CV #cs.LG
paper · pdf · doi:10.48550/arxiv.1411.4491
Paper is under consideration at Pattern Recognition Letters
openalex publication_date 2014/11/17 · arxiv created 2015/04/29 · arxiv updated 2015/04/30 · openalex created_date 2022/10/01 · openalex updated_date 2026/07/28
Domain adaptation aims at adapting the knowledge acquired on a source domain to a new different but related target domain. Several approaches have beenproposed for classification tasks in the unsupervised scenario, where no labeled target data are available. Most of the attention has been dedicated to searching a new domain-invariant representation, leaving the definition of the prediction function to a second stage. Here we propose to learn both jointly. Specifically we learn the source subspace that best matches the target subspace while at the same time minimizing a regularized misclassification loss. We provide an alternating optimization technique based on stochastic sub-gradient descent to solve the learning problem and we demonstrate its performance on several domain adaptation tasks.