Conference Paper (https://uobevents.eventsair.com/cikm2023//)
Title: CDR: Conservative Doubly Robust Learning for Debiased Recommendation
Please refer to the python codes in the "CDR" folder for Conservative Doubly Robust experiments.
Please refer to the python codes in the "baseline" folder for baseline experiments.
Coat and Yahoo datasets are already included in this project.
KuaiRand dataset can be download at (https://kuairand.com/). You can preprocess the data using python script in data/kuairand.
Our experimental environment is shown below:
python version: 3.7.12
pytorch version: 1.9.0
cuda version: 10.2
numpy version: 1.21.6
We follow the previous study, which is shown below:
@inproceedings{li2023stabledr,
title={StableDR: Stabilized Doubly Robust Learning for Recommendation on Data Missing Not at Random},
author={Li, Haoxuan and Zheng, Chunyuan and Wu, Peng},
booktitle={The Eleventh International Conference on Learning Representations},
year={2023}
}