https://github.com/bluer555/KernelGCN

Codes for NIPS 2019 Paper: Rethinking Kernel Methods for Node Representation Learning on Graphs
https://github.com/bluer555/KernelGCN

Last synced: 1 day ago
JSON representation

Codes for NIPS 2019 Paper: Rethinking Kernel Methods for Node Representation Learning on Graphs

• Host: GitHub
• URL: https://github.com/bluer555/KernelGCN
• Owner: bluer555
• Created: 2019-09-07T03:14:55.000Z (about 5 years ago)
• Default Branch: master
• Last Pushed: 2020-02-20T22:05:24.000Z (over 4 years ago)
• Last Synced: 2024-08-03T15:17:16.726Z (3 months ago)
• Language: Python
• Homepage:
• Size: 7.36 MB
• Stars: 33
• Watchers: 4
• Forks: 8
• Open Issues: 3
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

        
# Rethinking Kernel Methods for Node Representation Learning on Graphs

Training code for the paper

**[Rethinking Kernel Methods for Node Representation Learning on Graphs]

(https://arxiv.org/pdf/1910.02548.pdf)**, NIPS 2019

## Overview

We present a novel theoretical kernel-based framework for node classification. Our approach is motivated by graph kernel methodology but extended to learn the node representations capturing the structural information in a graph. We theoretically show that our formulation is as powerful as any positive semidefinite kernels. Our framework is flexible and complementary to other graph-based deep learning models, e.g., Graph Convolutional Networks (GCNs).



### Prerequisites

This package has the following requirements:

* `Python 3.6`

* `Pytorch 0.4.1`

* `numpy`

* `scipy`

* `networkx`

## Training

python train.py

## Citation

If you find this code useful in your research, please consider citing:

```

@inproceedings{tian2019rethinking,

  title={Rethinking kernel methods for node representation learning on graphs},

  author={Tian, Yu and Zhao, Long and Peng, Xi and Metaxas, Dimitris},

  booktitle={Advances in Neural Information Processing Systems},

  pages={11681--11692},

  year={2019}

}

```