This our official implementation via Pytorch for the paper:
Xiaoyu Du, Zike Wu, Fuli Feng, Xiangnan He and Jinhui Tang. Invariant Representation Learning for Multimedia Recommendation. In ACM MM`22, October 10–14, 2022, Lisboa, Portugal.
- Python==3.8.10
- Pytorch==1.11.0+cu113
- numpy, scipy, argparse, logging, sklearn, tqdm
First, we run UltraGCN for pretraining:
python main.py --model UltraGCN
After that, we run the InvRL model:
python main.py --model InvRL