A survey on heterogeneous transfer learning

Journal of Big Data

Table 3 Surveyed HTL methods which require no target labels

Methods	Characteristics	Sections
CT-SVM [70]	Symmetric, CCA, transfer SVM	“CT-SVM”
HHTL [71]	Asymmetric, requires source-target correspondence data, deep learning, mSDA	“HHTL”
HDCC [76]	Symmetric, CCA, group-weighing, video annotation, multi-source	“HDCC”
CL-SCL [78]	Symmetric, structural correspondence learning, text classification	“CL-SCL”
HDP [80]	Asymmetric through metric selection and matching	“HDP”
FuzzyTL [85]	Fuzzy logic, intelligent environments, FIS	“FuzzyTL”
UFSR [12]	Asymmetric, domain-dependent metafeatures to compute similarity	“FSR (IFSR, UFSR, ELFSR)”
ELFSR [12]	Asymmetric, FSR ensemble, multi-source, voting method	“FSR (IFSR, UFSR, ELFSR)”
RLG, GLG [11]	Symmetric, LMM, Grassmann manifold	“RLG, GLG”