Official implementation of our AAAI 2025 Oral paper:
"Behavior Importance-Aware Graph Neural Architecture Search for Cross-Domain Recommendation"
Feel free to star ⭐ this repo if you find it helpful!

conda create -n bignas python=3.10 -y
conda activate bignas

conda install pytorch torchvision torchaudio pytorch-cuda=12.4 -c pytorch -c nvidia

pip install torch_geometric
pip install pyg_lib torch_scatter torch_sparse torch_cluster torch_spline_conv \
     -f https://data.pyg.org/whl/torch-2.6.0+cu124.html

pip install -r requirements.txt

We provide an example of how to run the full BiGNAS training and evaluation pipeline.
Please simply run:

bash run.sh

The dataset will be automatically downloaded.
You can easily replace the --categories and --target arguments in the script to reproduce results for other task pairs.

If you use this code or find our work useful, please cite:

@article{ge2025behavior,
  title   = {Behavior Importance-Aware Graph Neural Architecture Search for Cross-Domain Recommendation},
  author  = {Ge, Chendi and Wang, Xin and Zhang, Ziwei and Qin, Yijian and Chen, Hong and Wu, Haiyang and Zhang, Yang and Yang, Yuekui and Zhu, Wenwu},
  journal = {Proceedings of the AAAI Conference on Artificial Intelligence},
  volume  = {39},
  number  = {11},
  pages   = {11708--11716},
  year    = {2025}
}

If you have any questions, feel free to open an issue or contact the first author at gcd23@mails.tsinghua.edu.cn.

This project is licensed under the MIT License.
See the LICENSE file for details.