KerTL

Code accompanying the paper Cross-Domain Kernel Induction for Transfer Learning.

Prerequisites

- Python, NumPy, Scipy, scikit-learn
- CVXOPT

Usage

To reproduce KerTL result in Table 2 of our paper, run

    $ python run_table2.py

To reproduce KerTL result in Table 3 of our paper, run

    $ python run_table3.py

Dataset

Use

$ 7z x mnist.7z -o.
$ 7z x apr.7z -o.

to decompress the data into the current folder.

More Info

This repository is by Wei-Cheng Chang, Yuexin Wu, Hanxiao Liu, Yiming Yang, and contains the source code to reproduce the experiments in our paper Cross-Domain Kernel Induction for Transfer Learning. If you find this repository helpful in your publications, please consider citing our paper.

@inproceedings{chang2017cross,
    title={Cross-Domain Kernel Induction for Transfer Learning.},
    author={Chang, Wei-Cheng and Wu, Yuexin and Liu, Hanxiao and Yang, Yiming},
    booktitle={AAAI},
    pages={1763--1769},
    year={2017}
}

For any questions and comments, please send your email to [email protected] or [email protected]