Command Palette
Search for a command to run...
3 months ago
GIDN: A Lightweight Graph Inception Diffusion Network for High-efficient Link Prediction
Zixiao Wang Yuluo Guo Jin Zhao Yu Zhang Hui Yu Xiaofei Liao Biao Wang Ting Yu
Abstract
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.
Benchmarks
| Benchmark | Methodology | Metrics |
|---|---|---|
| link-property-prediction-on-ogbl-collab | GIDN@YITU | Ext. data: No Number of params: 60449025 Test Hits@50: 0.7096 ± 0.0055 Validation Hits@50: 0.9620 ± 0.0040 |
| link-property-prediction-on-ogbl-ddi | GIDN@YITU | Ext. data: No Number of params: 3506691 Test Hits@20: 0.9542 ± 0.0000 Validation Hits@20: 0.8258 ± 0.0000 |
Build AI with AI
From idea to launch — accelerate your AI development with free AI co-coding, out-of-the-box environment and best price of GPUs.
AI Co-coding
Ready-to-use GPUs
Best Pricing
Hyper Newsletters
Subscribe to our latest updates
We will deliver the latest updates of the week to your inbox at nine o'clock every Monday morning
Powered by MailChimp