Skip to main content

Advertisement

Table 4 Heterogeneous transfer learning approaches surveyed in “Heterogeneous transfer learning” section listing various characteristics of each approach

From: A survey of transfer learning

Approach Transfer category Source data Target data Multiple sources Generic solution Negative transfer
CLSCL [91] Symmetric feature Labeled Unlabeled    
HeMap [105] Symmetric feature Labeled Limited labels    
DAMA [121] Symmetric feature Labeled Limited labels  
HTLIC [146] Symmetric feature Unlabeled Abundant labels    
TTI [92] Symmetric feature Labeled Limited labels    
HFA [30] Symmetric feature Labeled Limited labels    
SHFA [64] Symmetric feature Labeled Limited labels    
ARC-t [58] Asymmetric feature Labeled Limited labels    
MOMAP [46] Asymmetric feature Labeled Limited labels   
SHFR [145] Asymmetric feature Labeled Limited labels    
HDP [77] Asymmetric feature Labeled Unlabeled    
HHTL [144] Asymmetric feature Labeled Unlabeled