Learning Linear and Nonlinear Low-Rank Structure in Multi-Task Learning

Prerequisites

• Python3
• PyTorch ==1.6.0
• Numpy

Datasets

• fc7 layer feature:
  • office_31_fc7.txt
  • image_clef_fc7.txt
  • office_Home_fc7.txt
  • office-catech_fc7.txt
  • DomainNet_fc7.txt
• pool5 layer feature:
  • office_31_pool5.txt
  • image_clef_pool5.txt
  • office_Home_pool5.txt
  • office-catech_pool5.txt
  • DomainNet_pool5.txt

Download the pre-processed fc7 layer feature and pool5 layer feature from below link.

https://drive.google.com/drive/folders/1mAc2MPMIzChruQ6SBUC1eHB3XLSaDXvK?usp=sharing

Training

1. Downloading the dataset(s) from above link.

2. Run the experiment(s) (task fc7 layer as example):

   cd fc7
   python main.py

   To run this experiment, you need to modify the path datafile = 'xxx.txt' before you run python main.py.

Citation

If you use this code for your research, please consider citing:

@inproceedings{zhang2021multi,
  title={Multi-task learning via generalized tensor trace norm},
  author={Zhang, Yi and Zhang, Yu and Wang, Wei},
  booktitle={Proceedings of the 27th ACM SIGKDD Conference on Knowledge Discovery \& Data Mining},
  pages={2254--2262},
  year={2021}
}

@article{zhang2022learning,
  title={Learning Linear and Nonlinear Low-Rank Structure in Multi-Task Learning},
  author={Zhang, Yi and Zhang, Yu and Wang, Wei},
  journal={IEEE Transactions on Knowledge and Data Engineering},
  year={2022},
  publisher={IEEE}
}

Contact

If you have any problem about our code, feel free to contact [email protected].