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

2022-10-04Link Prediction

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.

Results

TaskDatasetMetricValueModel
Link Property Predictionogbl-ddiNumber of params3506691GIDN@YITU
Link Property Predictionogbl-collabNumber of params60449025GIDN@YITU

Related Papers