
Unsupervised Hierarchical Graph Representation Learning by Mutual Information Maximization
Graph representation learning based on graph neural networks (GNNs) can ...
read it

SelfSupervised Graph Representation Learning via Global Context Prediction
To take full advantage of fastgrowing unlabeled networked data, this pa...
read it

HDMI: Highorder Deep Multiplex Infomax
Networks have been widely used to represent the relations between object...
read it

node2coords: Graph Representation Learning with Wasserstein Barycenters
In order to perform network analysis tasks, representations that capture...
read it

Learning Robust Representations with Graph Denoising Policy Network
Graph representation learning, aiming to learn lowdimensional represent...
read it

NodeCentric Graph Learning from Data for Brain State Identification
Datadriven graph learning models a network by determining the strength ...
read it

Graphs in machine learning: an introduction
Graphs are commonly used to characterise interactions between objects of...
read it
Graph InfoClust: Leveraging clusterlevel node information for unsupervised graph representation learning
Unsupervised (or selfsupervised) graph representation learning is essential to facilitate various graph data mining tasks when external supervision is unavailable. The challenge is to encode the information about the graph structure and the attributes associated with the nodes and edges into a low dimensional space. Most existing unsupervised methods promote similar representations across nodes that are topologically close. Recently, it was shown that leveraging additional graphlevel information, e.g., information that is shared among all nodes, encourages the representations to be mindful of the global properties of the graph, which greatly improves their quality. However, in most graphs, there is significantly more structure that can be captured, e.g., nodes tend to belong to (multiple) clusters that represent structurally similar nodes. Motivated by this observation, we propose a graph representation learning method called Graph InfoClust (GIC), that seeks to additionally capture clusterlevel information content. These clusters are computed by a differentiable Kmeans method and are jointly optimized by maximizing the mutual information between nodes of the same clusters. This optimization leads the node representations to capture richer information and nodal interactions, which improves their quality. Experiments show that GIC outperforms stateofart methods in various downstream tasks (node classification, link prediction, and node clustering) with a 0.9 over the best competing approach, on average.
READ FULL TEXT
Comments
Costas Mavromatis ∙
arxiv: https://arxiv.org/abs/2009.06946
github: https://github.com/cmavro/GraphInfoClustGIC
extras: https://github.com/cmavro/awesomeunsupervisedgnns
∙ reply