Source code for GNN-LSPE (Graph Neural Networks with Learnable Structural and Positional Representations)

Overview

Graph Neural Networks with
Learnable Structural and Positional Representations

Source code for the paper "Graph Neural Networks with Learnable Structural and Positional Representations" by Vijay Prakash Dwivedi, Anh Tuan Luu, Thomas Laurent, Yoshua Bengio and Xavier Bresson.

We propose a novel GNN architecture in which the structural and positional representations are decoupled, and are learnt separately to learn these two essential properties. The architecture, named MPGNNs-LSPE (MPGNNs with Learnable Structural and Positional Encodings), is generic that it can be applied to any GNN model of interest which fits into the popular 'message-passing framework', including Transformers.

1. Repo installation

Follow these instructions to install the repo and setup the environment.

2. Download datasets

Proceed as follows to download the benchmark datasets.

3. Reproducibility

Use this page to run the codes and reproduce the published results.

4. Reference

📃 Paper on arXiv

@article{dwivedi2021graph,
  title={Graph Neural Networks with Learnable Structural and Positional Representations},
  author={Dwivedi, Vijay Prakash and Luu, Anh Tuan and Laurent, Thomas and Bengio, Yoshua and Bresson, Xavier},
  journal={arXiv preprint arXiv:2110.07875},
  year={2021}
}

This is an open-source toolkit for Heterogeneous Graph Neural Network(OpenHGNN) based on DGL [Deep Graph Library] and PyTorch.

519 Jan 2, 2023

Code for "Learning Structural Edits via Incremental Tree Transformations" (ICLR'21)

Learning Structural Edits via Incremental Tree Transformations Code for "Learning Structural Edits via Incremental Tree Transformations" (ICLR'21) 1.

40 Dec 23, 2022

Some tentative models that incorporate label propagation to graph neural networks for graph representation learning in nodes, links or graphs.

1 Nov 18, 2021

Comments

Avoid h being overwritten by residual p

Hi,

Congratulations for the nice paper&repo!

I think I found a small typo that could influence the performance of running PNA with residual=True, see attached.

Best, Andreea

opened by andreeadeac22 3
Migrate to partial node classification tasks

Dear Author,

Thanks so much for the code!

I don't know if LSPE can be used for sub-graph node classification (prediction on part of the nodes). I checked another repo you wrote (https://github.com/graphdeeplearning/benchmarking-gnns) and it seems I have to generate the features of all the nodes/edges first.

opened by ChengFu0118 1

Source code for GNN-LSPE (Graph Neural Networks with Learnable Structural and Positional Representations)

Related tags

Overview

Graph Neural Networks with
Learnable Structural and Positional Representations

1. Repo installation

2. Download datasets

3. Reproducibility

4. Reference

You might also like...

Learnable Motion Coherence for Correspondence Pruning

Sparse R-CNN: End-to-End Object Detection with Learnable Proposals, CVPR2021

Leibniz is a python package which provide facilities to express learnable partial differential equations with PyTorch

A PyTorch implementation of "Graph Classification Using Structural Attention" (KDD 2018).

Source code of NeurIPS 2021 Paper ''Be Confident! Towards Trustworthy Graph Neural Networks via Confidence Calibration''

Code for the TIP 2021 Paper "Salient Object Detection with Purificatory Mechanism and Structural Similarity Loss"

This is an open-source toolkit for Heterogeneous Graph Neural Network(OpenHGNN) based on DGL [Deep Graph Library] and PyTorch.

Code for "Learning Structural Edits via Incremental Tree Transformations" (ICLR'21)

Some tentative models that incorporate label propagation to graph neural networks for graph representation learning in nodes, links or graphs.

Comments

Avoid h being overwritten by residual p

Migrate to partial node classification tasks

Owner

Vijay Prakash Dwivedi

DIR-GNN - Discovering Invariant Rationales for Graph Neural Networks

A static analysis library for computing graph representations of Python programs suitable for use with graph neural networks.

A graph neural network (GNN) model to predict protein-protein interactions (PPI) with no sample features

The source code of the paper "Understanding Graph Neural Networks from Graph Signal Denoising Perspectives"

[CVPR 2021] A Peek Into the Reasoning of Neural Networks: Interpreting with Structural Visual Concepts

TensorFlow code for the neural network presented in the paper: "Structural Language Models of Code" (ICML'2020)

Relative Positional Encoding for Transformers with Linear Complexity

PyTorch implementation of Rethinking Positional Encoding in Language Pre-training

Incremental Transformer Structure Enhanced Image Inpainting with Masking Positional Encoding (CVPR2022)

Learnable Multi-level Frequency Decomposition and Hierarchical Attention Mechanism for Generalized Face Presentation Attack Detection