GIDN: A Lightweight Graph Inception Diffusion Network for High-efficient Link Prediction
2022-10-04Unverified0· sign in to hype
Zixiao Wang, Yuluo Guo, Jin Zhao, Yu Zhang, Hui Yu, Xiaofei Liao, Biao Wang, Ting Yu
Unverified — Be the first to reproduce this paper.
ReproduceAbstract
In this paper, we propose a Graph Inception Diffusion Networks(GIDN) model. This model generalizes graph diffusion in different feature spaces, and uses the inception module to avoid the large amount of computations caused by complex network structures. We evaluate GIDN model on Open Graph Benchmark(OGB) datasets, reached an 11% higher performance than AGDN on ogbl-collab dataset.
Tasks
Benchmark Results
| Dataset | Model | Metric | Claimed | Verified | Status |
|---|---|---|---|---|---|
| ogbl-collab | GIDN@YITU | Number of params | 60,449,025 | — | Unverified |
| ogbl-ddi | GIDN@YITU | Number of params | 3,506,691 | — | Unverified |