Abstract:Graphs and networks are very common data structure for modelling complex systems that are composed of a number of nodes and topologies, such as social networks, citation networks, biological protein-protein interactions networks, etc. In recent years, machine learning has become an efficient technique to obtain representation of graph for downstream graph analysis tasks, including node classification, link prediction, and community detection. Different with traditional graph analytical models, the representation learning on graph tries to learn low dimensional embeddings by means of machine learning models that could be trained in supervised, unsupervised or semi-supervised manners. Compared with traditional approaches that directly use input node attributes, these embeddings are much more informative and helpful for graph analysis. There are a number of developed models in this respect, that are different in the ways of measuring similarity of vertexes in both original space and feature space. In order to learn more efficient node representation with better generalization property, we propose a task-independent graph representation model, called as graph deconvolutional network (GDN), and corresponding unsupervised learning algorithm in this paper. Different with graph convolution network (GCN) from the scratch, which produces embeddings by convolving input attribute vectors with learned filters, the embeddings of the proposed GDN model are desired to be convolved with filters so that reconstruct the input node attribute vectors as far as possible. The embeddings and filters are alternatively optimized in the learning procedure. The correctness of the proposed GDN model is verified by multiple tasks over several datasets. The experimental results show that the GDN model outperforms existing alternatives with a big margin.

Visualizing Graph Neural Networks with CorGIE: Corresponding a Graph to Its Embedding

CoRe-GD: A Hierarchical Framework for Scalable Graph Visualization with GNNs

GNN 101: Visual Learning of Graph Neural Networks in Your Web Browser

Graph Neural Networks Inspired by Classical Iterative Algorithms

Explainable, Stable, and Scalable Graph Convolutional Networks for Learning Graph Representation

Geodesic Graph Neural Network for Efficient Graph Representation Learning

Graphion: graph neural network-based visual correlation feature extraction module

Isometric Graph Neural Networks

Unveiling Global Interactive Patterns across Graphs: Towards Interpretable Graph Neural Networks

Graph deconvolutional networks.

Learning the Network of Graphs for Graph Neural Networks

Feature Correlation Aggregation: on the Path to Better Graph Neural Networks

A data-centric framework of improving graph neural networks for knowledge graph embedding

Commute Graph Neural Networks

Cooperative Explanations of Graph Neural Networks.

Graph Neural Networks: Methods, Applications, and Opportunities

Local Clustering Graph Neural Networks

KerGNNs: Interpretable Graph Neural Networks with Graph Kernels

Fisher Information Embedding for Node and Graph Learning

Graph in Graph Neural Network