# A Generalization of Convolutional Neural Networks to Graph-Structured Data

arXiv: https://arxiv.org/abs/1704.08165 · HTML: https://mlsanity.com/paper/1704.08165/

## Key facts

- Cited by 9 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 26 Apr 2017.

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

- [Adaptive Filters in Graph Convolutional Neural Networks](https://mlsanity.com/paper/2105.10377/) (2021)
- [Spatial-temporal traffic modeling with a fusion graph reconstructed by tensor decomposition](https://mlsanity.com/paper/2212.05653/) (2022)
- [Locally temporal-spatial pattern learning with graph attention mechanism for EEG-based emotion recognition](https://mlsanity.com/paper/2208.11087/) (2022)
- [Neural Algorithmic Reasoners informed Large Language Model for Multi-Agent Path Finding](https://mlsanity.com/paper/2508.17971/) (2025)
- [Prediction of Hereditary Cancers Using Neural Networks](https://mlsanity.com/paper/2106.13682/) (2021)
- [Predicting vacant parking space availability zone-wisely: a graph based spatio-temporal prediction approach](https://mlsanity.com/paper/2205.02113/) (2022)
- [ESGCN: Edge Squeeze Attention Graph Convolutional Network for Traffic Flow Forecasting](https://mlsanity.com/paper/2307.01227/) (2023)
- [Global-Aware Enhanced Spatial-Temporal Graph Recurrent Networks: A New Framework For Traffic Flow Prediction](https://mlsanity.com/paper/2401.04135/) (2024)
- [Graph Neural Networks in Wind Power Forecasting](https://mlsanity.com/paper/2507.00105/) (2025)

## Abstract (arXiv)

This paper introduces a generalization of Convolutional Neural Networks (CNNs) from low-dimensional grid data, such as images, to graph-structured data. We propose a novel spatial convolution utilizing a random walk to uncover the relations within the input, analogous to the way the standard convolution uses the spatial neighborhood of a pixel on the grid. The convolution has an intuitive interpretation, is efficient and scalable and can also be used on data with varying graph structure. Furthermore, this generalization can be applied to many standard regression or classification problems, by learning the the underlying graph. We empirically demonstrate the performance of the proposed CNN on MNIST, and challenge the state-of-the-art on Merck molecular activity data set.