Supervised transfer learning algorithms utilize labeled data from auxiliary domains for learning in another domain where labeled data is scarce or absent. Given sufficient cross-domain corresponding instances, one can ...