Pytorch implementation for the paper: Breaking Isolation: Multimodal Graph Fusion for Multimedia Recommendation by Edge-wise Modulation, ACM Mulimedia 2022.

The code has been tested running under Python 3.8.5. The required packages are as follows:

• Pytorch 1.7.1
• CUDA 11.3
• torch-geometric 1.7.2

Download data following the instructions here.

• Movielens dataset (EgoGCN-hard)

python main.py --l_r=0.0002 --weight_decay=0.0001 --weight_mode=GCN --num_routing=3 --data_path=movielens --batch_size 2048 --alpha_threshold='>0.4' --reattn=True --central=central_item

• Tiktok dataset (EgoGCN-hard)

python main.py --l_r=0.0003 --weight_decay=0.001 --weight_mode=GCN --num_routing=1 --data_path=Tiktok --batch_size 1024 --alpha_threshold '>0.6' --reattn=True --central=central_item

For EgoGCN-soft, simply setting --alpha_threshold '=0'

For other variants in the paper, for instance, EgoGCN-NA,

python main.py --l_r=0.0002 --weight_decay=0.0001 --weight_mode=GCN --num_routing=3 --data_path=movielens --batch_size 2048 --alpha_threshold='>99' --reattn=True --central=central_item --NA

python main.py --l_r=0.0003 --weight_decay=0.001 --weight_mode=GCN --num_routing=1 --data_path=Tiktok --batch_size 1024 --alpha_threshold '>99' --reattn=True --central=central_item --NA

If you found this code useful, please cite the following paper:

@inproceedings{DBLP:conf/mm/ChenWWZS22,
  author    = {Feiyu Chen and
               Junjie Wang and
               Yinwei Wei and
               Hai{-}Tao Zheng and
               Jie Shao},
  title     = {Breaking Isolation: Multimodal Graph Fusion for Multimedia Recommendation
               by Edge-wise Modulation},
  booktitle = {{MM} '22: The 30th {ACM} International Conference on Multimedia, Lisboa,
               Portugal, October 10 - 14, 2022},
  pages     = {385--394},
  publisher = {{ACM}},
  year      = {2022},
}