A demo application running on the wireless data set, this is a kernelized version.
Knowledge Preserving and Distribution Alignment for Heterogeneous Domain Adaptation, ACM Transactions on Information Systems, 2022
Authors: Hanrui Wu, Qingyao Wu, Michael K. Ng
Please note that the initialization of the mapping matrices P and Q has an impact on the result.
Besides, the parameters in the demo file are selected from different sets of values.
Detailed information on the parameter selection refers to the paper.
Data can be downloaded at [data].

The codes are written based on the ManOpt provided by Z. Wen and W. Yin. A feasible method for optimization with orthogonality constraints. Mathematical Programming 2013

Please cite our paper if you find this code helpful to your research:
@article{wu2022knowledge,
title = {Knowledge Preserving and Distribution Alignment for Heterogeneous Domain Adaptation},
author = {Wu, Hanrui and Wu, Qingyao and Ng, Michael Kwok-Po},
journal = {ACM Transactions on Information Systems (TOIS)},
year = {2022},
volume={40},
number={1},
pages={1--29},
year={2021},
publisher={ACM New York, NY}
}