Hierarchical Graph Neural Network Based on Semi-Implicit Variational Inference

Graph neural network (GNN) has obtained outstanding achievements in relational data. However, these data have uncertain properties, for example, spurious edges may be included. Recently, variational graph autoencoder (VGAE) has been proposed to solve this problem. However, the distributional assumpt...

Full description

Saved in:
Bibliographic Details
Published in:IEEE transactions on cognitive and developmental systems 2023-06, Vol.15 (2), p.887-895
Main Authors: Su, Hai-Long, Li, Zhi-Peng, Zhu, Xiao-Bo, Yang, Li-Na, Gribova, Valeriya, Filaretov, Vladimir Fedorovich, Cohn, Anthony G., Huang, De-Shuang
Format: Article
Language:English
Subjects:
Datasets
Decoding
Flexibility
Gaussian distribution
Graph neural network (GNN)
Graph neural networks
Graph theory
Graphite
hierarchical frame
Inference
Iterative methods
latent variable
Neural networks
Proteins
Random variables
semi-implicit model
Uncertainty
variation inference
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Graph neural network (GNN) has obtained outstanding achievements in relational data. However, these data have uncertain properties, for example, spurious edges may be included. Recently, variational graph autoencoder (VGAE) has been proposed to solve this problem. However, the distributional assumptions in the variational family restrict the variational inference (VI) flexibility and they define variational families using mean field, which can not capture complex posterior distributional. To solve the above question, in this article, we proposed a novel GNN model based on semi-implicit VI (SIVI), which can embed the node to the latent space to improve VI flexibility and enhance VI expressiveness with mixing distribution. Specifically, to approximate the true posterior, a variational posterior was given utilizing a semi-implicit hierarchical variational framework, which can model complex posterior. Moreover, an iterative decoder is used to better capture graph properties. Besides, due to the hierarchical structure in our model, it can incorporation neighbor information between nodes. Experiments on multiple data sets, Our method has achieved state-of-the-art results compared to other similar methods. Particularly, on the citation data set Citeseer without features, our method outperforms VGAE by 9%.
ISSN:2379-8920
2379-8939
DOI:10.1109/TCDS.2022.3193398