Skip to content

Latest commit

 

History

History
26 lines (17 loc) · 969 Bytes

README.md

File metadata and controls

26 lines (17 loc) · 969 Bytes

IJCAI'18 - AdaTTL

MATLAB code for paper Does Tail Label Help for Large-Scale Multi-Label Learning?

The package includes the MATLAB code of the Adaptively Trimming off Tail Labels (AdaTTL) which aims to filter out tail labels and facilitate fast prediction as well as compact model size for large-scale multi-label learning.

You will find an example of using this code in script 'example.m'.

Please cite this paper when using this code for your research.

@inproceedings{DBLP:conf/ijcai/WeiL18,
  title={Does Tail Label Help for Large-Scale Multi-Label Learning},
  author={Wei, Tong and Li, Yu-Feng},
  booktitle={Proceedings of the 27th International Joint Conference on Artificial Intelligence},
  pages={2847--2853},
  year={2018},
  address={Stockholm, Sweden}
}

For any problem concerning the code, please feel free to contact Tong Wei at weit@lamda.nju.edu.cn

Last updated on May 15, 2018