TASK	DATASET	MODEL	METRIC NAME	METRIC VALUE	GLOBAL RANK
Graph Classification	20NEWS	sKNN-LDS	Accuracy	47.9	# 1
Graph Classification	Cancer	sKNN-LDS	Accuracy	95.7	# 1
Graph Classification	Citeseer	sKNN-LDS	Accuracy	73.7	# 1
Graph Classification	COLLAB	sGIN	Accuracy	80.71%	# 10
Graph Classification	Cora	sKNN-LDS	Accuracy	72.3	# 1
Graph Classification	Digits	sKNN-LDS	Accuracy	92.5	# 1
Graph Classification	IMDb-B	sGIN	Accuracy	77.94%	# 6
Graph Classification	IMDb-M	sGIN	Accuracy	54.52%	# 7
Graph Classification	MUTAG	sGIN	Accuracy	94.14%	# 8
Graph Classification	NCI1	sGIN	Accuracy	83.85%	# 17
Graph Classification	PROTEINS	sGIN	Accuracy	78.97%	# 10
Graph Classification	PTC	sGIN	Accuracy	73.56%	# 5
Link Prediction	Pubmed	sGraphite-VAE	AUC	94.8%	# 8
Link Prediction	Pubmed	sGraphite-VAE	AP	96.3%	# 7
Graph Classification	Wine	sKNN-LDS	Accuracy	98	# 1

Badge	Markdown
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/neighborhood-enlargement-in-graph-neural/graph-classification-on-20news)](https://paperswithcode.com/sota/graph-classification-on-20news?p=neighborhood-enlargement-in-graph-neural)`
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/neighborhood-enlargement-in-graph-neural/graph-classification-on-cancer)](https://paperswithcode.com/sota/graph-classification-on-cancer?p=neighborhood-enlargement-in-graph-neural)`
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/neighborhood-enlargement-in-graph-neural/graph-classification-on-citeseer)](https://paperswithcode.com/sota/graph-classification-on-citeseer?p=neighborhood-enlargement-in-graph-neural)`
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/neighborhood-enlargement-in-graph-neural/graph-classification-on-cora)](https://paperswithcode.com/sota/graph-classification-on-cora?p=neighborhood-enlargement-in-graph-neural)`
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/neighborhood-enlargement-in-graph-neural/graph-classification-on-digits)](https://paperswithcode.com/sota/graph-classification-on-digits?p=neighborhood-enlargement-in-graph-neural)`
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/neighborhood-enlargement-in-graph-neural/graph-classification-on-wine)](https://paperswithcode.com/sota/graph-classification-on-wine?p=neighborhood-enlargement-in-graph-neural)`
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/neighborhood-enlargement-in-graph-neural/graph-classification-on-ptc)](https://paperswithcode.com/sota/graph-classification-on-ptc?p=neighborhood-enlargement-in-graph-neural)`
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/neighborhood-enlargement-in-graph-neural/graph-classification-on-imdb-b)](https://paperswithcode.com/sota/graph-classification-on-imdb-b?p=neighborhood-enlargement-in-graph-neural)`
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/neighborhood-enlargement-in-graph-neural/graph-classification-on-imdb-m)](https://paperswithcode.com/sota/graph-classification-on-imdb-m?p=neighborhood-enlargement-in-graph-neural)`
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/neighborhood-enlargement-in-graph-neural/graph-classification-on-mutag)](https://paperswithcode.com/sota/graph-classification-on-mutag?p=neighborhood-enlargement-in-graph-neural)`
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/neighborhood-enlargement-in-graph-neural/link-prediction-on-pubmed)](https://paperswithcode.com/sota/link-prediction-on-pubmed?p=neighborhood-enlargement-in-graph-neural)`
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/neighborhood-enlargement-in-graph-neural/graph-classification-on-collab)](https://paperswithcode.com/sota/graph-classification-on-collab?p=neighborhood-enlargement-in-graph-neural)`
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/neighborhood-enlargement-in-graph-neural/graph-classification-on-proteins)](https://paperswithcode.com/sota/graph-classification-on-proteins?p=neighborhood-enlargement-in-graph-neural)`
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/neighborhood-enlargement-in-graph-neural/graph-classification-on-nci1)](https://paperswithcode.com/sota/graph-classification-on-nci1?p=neighborhood-enlargement-in-graph-neural)`

Mutual Information Maximization in Graph Neural Networks

21 May 2019 · Xinhan Di, Pengqian Yu, Rui Bu, Mingchao Sun ·

A variety of graph neural networks (GNNs) frameworks for representation learning on graphs have been recently developed. These frameworks rely on aggregation and iteration scheme to learn the representation of nodes. However, information between nodes is inevitably lost in the scheme during learning. In order to reduce the loss, we extend the GNNs frameworks by exploring the aggregation and iteration scheme in the methodology of mutual information. We propose a new approach of enlarging the normal neighborhood in the aggregation of GNNs, which aims at maximizing mutual information. Based on a series of experiments conducted on several benchmark datasets, we show that the proposed approach improves the state-of-the-art performance for four types of graph tasks, including supervised and semi-supervised graph classification, graph link prediction and graph edge generation and classification.

PDF Abstract

Code

Add Remove Mark official

CODE-SUBMIT/Graph_Neighborhood_1 official

CODE-SUBMIT/Neighborhood-Enlargemen… official

Tasks

Add Remove

General Classification

Graph Classification

Link Prediction

Representation Learning

Datasets

Pubmed

Cora PROTEINS Citeseer IMDB-BINARY MUTAG COLLAB NCI1 IMDB-MULTI PTC

Digits

Wine

Results from the Paper

Edit

Ranked #1 on Graph Classification on Citeseer

Get a GitHub badge

Task	Dataset	Model	Metric Name	Metric Value	Global Rank	Benchmark
Graph Classification	20NEWS	sKNN-LDS	Accuracy	47.9	# 1	Compare
Graph Classification	Cancer	sKNN-LDS	Accuracy	95.7	# 1	Compare
Graph Classification	Citeseer	sKNN-LDS	Accuracy	73.7	# 1	Compare
Graph Classification	COLLAB	sGIN	Accuracy	80.71%	# 10	Compare
Graph Classification	Cora	sKNN-LDS	Accuracy	72.3	# 1	Compare
Graph Classification	Digits	sKNN-LDS	Accuracy	92.5	# 1	Compare
Graph Classification	IMDb-B	sGIN	Accuracy	77.94%	# 6	Compare
Graph Classification	IMDb-M	sGIN	Accuracy	54.52%	# 7	Compare
Graph Classification	MUTAG	sGIN	Accuracy	94.14%	# 8	Compare
Graph Classification	NCI1	sGIN	Accuracy	83.85%	# 17	Compare
Graph Classification	PROTEINS	sGIN	Accuracy	78.97%	# 10	Compare
Graph Classification	PTC	sGIN	Accuracy	73.56%	# 5	Compare
Link Prediction	Pubmed	sGraphite-VAE	AUC	94.8%	# 8	Compare
Link Prediction	Pubmed	sGraphite-VAE	AP	96.3%	# 7	Compare
Graph Classification	Wine	sKNN-LDS	Accuracy	98	# 1	Compare

Methods

Add Remove

No methods listed for this paper. Add relevant methods here

Edit Social Preview

Mutual Information Maximization in Graph Neural Networks

Code Edit Add Remove Mark official

Tasks Edit Add Remove

Datasets Edit

Results from the Paper Edit

Methods Edit Add Remove

Code

Add Remove Mark official

Tasks

Add Remove

Datasets

Results from the Paper

Edit

Methods

Add Remove