Skip to content

Latest commit

 

History

History
21 lines (17 loc) · 1.06 KB

README.md

File metadata and controls

21 lines (17 loc) · 1.06 KB

Kalman Filtering Attention (wip)

Implementation of the Kalman Filtering Attention proposed in Kalman Filtering Attention for User Behavior Modeling in CTR Prediction

Will use this repository as guidance. Looks like the core of Kalman filtering is just 5 lines of code.

Citations

@inproceedings{NEURIPS2020_68ce199e,
    author = {Liu, Hu and LU, Jing and Zhao, Xiwei and Xu, Sulong and Peng, Hao and Liu, Yutong and Zhang, Zehua and Li, Jian and Jin, Junsheng and Bao, Yongjun and Yan, Weipeng},
    booktitle = {Advances in Neural Information Processing Systems},
    editor = {H. Larochelle and M. Ranzato and R. Hadsell and M.F. Balcan and H. Lin},
    pages = {9228--9238},
    publisher = {Curran Associates, Inc.},
    title = {Kalman Filtering Attention for User Behavior Modeling in CTR Prediction},
    url = {https://proceedings.neurips.cc/paper_files/paper/2020/file/68ce199ec2c5517597ce0a4d89620f55-Paper.pdf},
    volume = {33},
    year = {2020}
}