Symmetric Feature-based Transfer Learning

Homogeneous symmetric transformation takes both the source feature space Xs and target feature space Xt and learns feature transformations as to project each onto a common subspace Xc for adaptation purposes. This derived subspace becomes a domain-invariant feature subspace to associate cross-domain data, and in effect, reduces marginal distribution differences.

References

Day, O., & Khoshgoftaar, T.M. (2017). A survey on heterogeneous transfer learning. Journal of Big Data, 4(1), 29. Link.