# Mutual Information Maximization in Graph Neural Networks

arXiv: https://arxiv.org/abs/1905.08509 · HTML: https://mlsanity.com/paper/1905.08509/

## Key facts

- Cited by 9 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 21 May 2019.

## Built on by (most cited first, 9 of 9)

- [Variational Graph Normalized Auto-Encoders](https://mlsanity.com/paper/2108.08046/) (2021)
- [Text Serialization and Their Relationship with the Conventional Paradigms of Tabular Machine Learning](https://mlsanity.com/paper/2406.13846/) (2024)
- [Graph Kernel Neural Networks](https://mlsanity.com/paper/2112.07436/) (2021)
- [Learning Compact Representations of Neural Networks using DiscriminAtive Masking (DAM)](https://mlsanity.com/paper/2110.00684/) (2021)
- [Supervised Contrastive Learning with Structure Inference for Graph Classification](https://mlsanity.com/paper/2203.07691/) (2022)
- [Mathematical Formalism for Memory Compression in Selective State Space Models](https://mlsanity.com/paper/2410.03158/) (2024)
- [AutoHEnsGNN: Winning Solution to AutoGraph Challenge for KDD Cup 2020](https://mlsanity.com/paper/2111.12952/) (2021)
- [Towards Causal Classification: A Comprehensive Study on Graph Neural Networks](https://mlsanity.com/paper/2401.15444/) (2024)
- [GNN-LoFI: a Novel Graph Neural Network through Localized Feature-based Histogram Intersection](https://mlsanity.com/paper/2401.09193/) (2024)

## Abstract (arXiv)

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.