Traditional machine learning algorithms operate under the assumption that learning for each new task starts from scratch, thus disregarding knowledge acquired in previous domains. Naturally, if the domains encountered during learning are related, this tabula rasa approach wastes both data and computational resources in developing hypotheses that could have potentially been recovered by simply slightly modifying previously acquired knowledge. The field of transfer learning (TL), which has witnessed substantial growth in recent years, develops methods that attempt to utilize previously acquired knowledge in a source domain in order to improve the efficiency and accuracy of learning in a new, but related, target domain [7,6,1].