Subscribe to the PwC Newsletter
Join the community, edit social preview.
Add a new code entry for this paper
Remove a code repository from this paper, mark the official implementation from paper authors, add a new evaluation result row.
| TASK | DATASET | MODEL | METRIC NAME | METRIC VALUE | GLOBAL RANK | REMOVE |
|---|---|---|---|---|---|---|
| Node Classification | Citeseer | PairE | Accuracy | 75.53 | # 21 | |
| 30% trainning, unsupervised with linear classfier | Cora | PairE | Micro-F1 | 86.51 | # 1 | |
| Node Classification | Cora: fixed 20 node per class | PairE | Micro F1 | 75.12 | # 1 | |
| Node Classification | DBLP | PairE | Micro F1 | 80.58 | # 2 | |
| Node Classification | Deezer Romania | PairE | Micro-F1 | 0.68 | # 1 | |
| Node Classification | PPI | PairE | Micro F1 | 94.83 | # 1 | |
| Node Classification | Pubmed | PairE | F1 | 88.57 | # 1 |
- EDGE CLASSIFICATION
- GRAPH EMBEDDING
- GRAPH REPRESENTATION LEARNING
- NODE CLASSIFICATION
- REPRESENTATION LEARNING
Remove a task
Add a method
- AUTOENCODER
Remove a method
- AUTOENCODER -
Include the markdown at the top of your GitHub README.md file to showcase the performance of the model.
Badges are live and will be dynamically updated with the latest ranking of this paper.
| Badge | Markdown |
|---|---|
Edit Datasets
Graph representation learning beyond node and homophily.
3 Mar 2022 · You Li , Bei Lin , Binli Luo , Ning Gui · Edit social preview
Unsupervised graph representation learning aims to distill various graph information into a downstream task-agnostic dense vector embedding. However, existing graph representation learning approaches are designed mainly under the node homophily assumption: connected nodes tend to have similar labels and optimize performance on node-centric downstream tasks. Their design is apparently against the task-agnostic principle and generally suffers poor performance in tasks, e.g., edge classification, that demands feature signals beyond the node-view and homophily assumption. To condense different feature signals into the embeddings, this paper proposes PairE, a novel unsupervised graph embedding method using two paired nodes as the basic unit of embedding to retain the high-frequency signals between nodes to support node-related and edge-related tasks. Accordingly, a multi-self-supervised autoencoder is designed to fulfill two pretext tasks: one retains the high-frequency signal better, and another enhances the representation of commonality. Our extensive experiments on a diversity of benchmark datasets clearly show that PairE outperforms the unsupervised state-of-the-art baselines, with up to 101.1\% relative improvement on the edge classification tasks that rely on both the high and low-frequency signals in the pair and up to 82.5\% relative performance gain on the node classification tasks.
Code Edit Add Remove Mark official
Tasks edit add remove, datasets edit.
Results from the Paper Edit
| Task | Dataset | Model | Metric Name | Metric Value | Global Rank | Result | Benchmark |
|---|---|---|---|---|---|---|---|
| Node Classification | Citeseer | PairE | Accuracy | 75.53 | # 21 | ||
| 30% trainning, unsupervised with linear classfier | Cora | PairE | Micro-F1 | 86.51 | # 1 | ||
| Node Classification | Cora: fixed 20 node per class | PairE | Micro F1 | 75.12 | # 1 | ||
| Node Classification | DBLP | PairE | Micro F1 | 80.58 | # 2 | ||
| Node Classification | Deezer Romania | PairE | Micro-F1 | 0.68 | # 1 | ||
| Node Classification | PPI | PairE | Micro F1 | 94.83 | # 1 | ||
| Node Classification | Pubmed | PairE | F1 | 88.57 | # 1 |
Methods Edit Add Remove
- DOI: 10.1109/TKDE.2022.3146270
- Corpus ID: 246377448
Graph Representation Learning Beyond Node and Homophily
- You Li , Bei Lin , +1 author Ning Gui
- Published in IEEE Transactions on… 3 March 2022
- Computer Science
Figures and Tables from this paper
13 Citations
Multi-view graph representation learning beyond homophily, graph neural networks for graphs with heterophily: a survey, pytorch-geometric edge – a library for learning representations of graph edges.
- Highly Influenced
Homophily-oriented Heterogeneous Graph Rewiring
From motif to path: connectivity and homophily, ms-gda: improving heterogeneous recipe representation via multinomial sampling graph data augmentation, curvagn: curvature-based adaptive graph neural networks for predicting protein-ligand binding affinity, a survey on graph construction for geometric deep learning in medicine: methods and recommendations, pytorch-geometric edge – a library for learning representations of graph edges, label-aware aggregation on heterophilous graphs for node representation learning, 40 references, beyond homophily in graph neural networks: current limitations and effective designs.
- Highly Influential
A Representation Learning Framework for Property Graphs
Inductive representation learning on large graphs, deep graph infomax, is a single embedding enough learning node representations that capture multiple social contexts, representation learning on graphs with jumping knowledge networks, inductive and unsupervised representation learning on graph structured objects, node2vec: scalable feature learning for networks, representation learning on graphs: methods and applications, related papers.
Showing 1 through 3 of 0 Related Papers
Published in IEEE Transactions on Knowledge and Data Engineering 2022
You Li Bei Lin Binli Luo Ning Gui
Node Classification Beyond Homophily: Towards a General Solution
New citation alert added.
This alert has been successfully added and will be sent to:
You will be notified whenever a record that you have chosen has been cited.
To manage your alert preferences, click on the button below.
New Citation Alert!
Please log in to your account
Information & Contributors
Bibliometrics & citations, view options, supplementary material.
- Das S Ferdous S Halappanavar M Serra E Pothen A Baeza-Yates R Bonchi F (2024) AGS-GNN: Attribute-guided Sampling for Graph Neural Networks Proceedings of the 30th ACM SIGKDD Conference on Knowledge Discovery and Data Mining 10.1145/3637528.3671940 (538-549) Online publication date: 25-Aug-2024 https://dl.acm.org/doi/10.1145/3637528.3671940
- Zhao H Chen A Sun X Cheng H Li J Baeza-Yates R Bonchi F (2024) All in One and One for All: A Simple yet Effective Method towards Cross-domain Graph Pretraining Proceedings of the 30th ACM SIGKDD Conference on Knowledge Discovery and Data Mining 10.1145/3637528.3671913 (4443-4454) Online publication date: 25-Aug-2024 https://dl.acm.org/doi/10.1145/3637528.3671913
- Yan Y Chen Y Chen H Xu M Das M Yang H Tong H Oh A Naumann T Globerson A Saenko K Hardt M Levine S (2023) From trainable negative depth to edge heterophily in graphs Proceedings of the 37th International Conference on Neural Information Processing Systems 10.5555/3666122.3669196 (70162-70178) Online publication date: 10-Dec-2023 https://dl.acm.org/doi/10.5555/3666122.3669196
- Show More Cited By
Index Terms
Computing methodologies
Machine learning
Machine learning approaches
Neural networks
Information systems
Information systems applications
Data mining
Theory of computation
Design and analysis of algorithms
Graph algorithms analysis
Recommendations
Node classification of graph neural networks based on graph degree-symmetry.
In order to accurately reflect the information of each node's neighbours and the topological relationship of the graph structure, the graph neural network node classification task aims to treat each node and all of its neighbouring nodes as a subgraph ...
Generalized heterophily graph data augmentation for node classification
Graph data augmentations have demonstrated remarkable performance on homophilic graph neural networks (GNNs). Nevertheless, when transferred to a heterophilic graph, these augmentations are less effective for GNN models and lead to reduced ...
Chaotic Neural Oscillators with Deep Graph Neural Network for Node Classification
Node classification is a pivotal task in spam detection, community identification, and social network analysis. Compared with traditional graph learning methods, Graph Neural Networks (GNN) show superior performance in prediction tasks, but ...
Information
Published in.
- General Chairs:
UC Santa Barbara, USA
UC Los Angeles, USA
- Program Chairs:
Carnegie Mellon University, USA
University of Athens, Greece
Google, USA
Alibaba DAMO Academy
- SIGMOD: ACM Special Interest Group on Management of Data
- SIGKDD: ACM Special Interest Group on Knowledge Discovery in Data
Association for Computing Machinery
New York, NY, United States
Publication History
Permissions, check for updates, author tags.
- graph data augmentation
- graph machine learning
- graph neural network
- node classification
- Research-article
Funding Sources
Acceptance rates, contributors, other metrics, bibliometrics, article metrics.
- 4 Total Citations View Citations
- 707 Total Downloads
- Downloads (Last 12 months) 707
- Downloads (Last 6 weeks) 49
- Zhou Q Ding K Liu H Tong H Frommholz I Hopfgartner F Lee M Oakes M Lalmas M Zhang M Santos R (2023) Learning Node Abnormality with Weak Supervision Proceedings of the 32nd ACM International Conference on Information and Knowledge Management 10.1145/3583780.3614950 (3584-3594) Online publication date: 21-Oct-2023 https://dl.acm.org/doi/10.1145/3583780.3614950
View options
View or Download as a PDF file.
View online with eReader .
Login options
Check if you have access through your login credentials or your institution to get full access on this article.
Full Access
Share this publication link.
Copying failed.
Share on social media
Affiliations, export citations.
- Please download or close your previous search result export first before starting a new bulk export. Preview is not available. By clicking download, a status dialog will open to start the export process. The process may take a few minutes but once it finishes a file will be downloadable from your browser. You may continue to browse the DL while the export process is in progress. Download
- Download citation
- Copy citation
We are preparing your search results for download ...
We will inform you here when the file is ready.
Your file of search results citations is now ready.
Your search export query has expired. Please try again.
Navigation Menu
Search code, repositories, users, issues, pull requests..., provide feedback.
We read every piece of feedback, and take your input very seriously.
Saved searches
Use saved searches to filter your results more quickly.
To see all available qualifiers, see our documentation .
- Notifications You must be signed in to change notification settings
"Graph Representation Learning Beyond Node and Homophily"
syvail/PairE-Graph-Representation-Learning-Beyond-Node-and-Homophily
Folders and files.
| Name | Name | |||
|---|---|---|---|---|
| 8 Commits | ||||
Repository files navigation
Paire-graph-representation-learning-beyond-node-and-homophily.
We provide the source code of paire. Our model implementation is based on keras, which allows the model to be trained by GPU.
Installation
- run the following code pip install -r requirements.txt
The source code is saved in the pair-embedding.ipynb file
To run "PairE" on Cora network and evaluate the learned representations on multi-label node classification task, run the following command in the home directory of this project:
The supported input format is an edgelist or an adjlist:
The graph is assumed to be undirected and unweighted by default. The model needs additional features, the supported feature input format is as follow ( feature_i should be a float number):
The output contains two dataframes: pair_embedding has |E| lines for a graph with |E| edges. Node_embedding has n lines for a graph with n nodes. The |E| lines are as follows:
The n lines are as follows:
where dim1, ... , dimd is the d -dimensional representation learned by PairE .
Train model :
If you want to evaluate the learned node representations, you can input the node labels. It will use a portion of nodes(default:10%、30%、50%、70%、90%) to train a classifier and calculate F1-score on the rest dataset.
The supported input label format is
evaluate on multi-class node classfication:
Embedding visualization.
We apply the dimensionality reduction method like t-SNE to the embedded visualization, and visualize the embedding of different data sets, where the colors of nodes represent the labels of nodes.
- Jupyter Notebook 100.0%
IMAGES
COMMENTS
A novel unsupervised graph embedding method, PairE, that uses two paired nodes as the basic unit to support node-related and edge-related tasks. The paper shows that PairE outperforms the state-of-the-art baselines on various benchmark datasets.
Extensive experiments on synthetic and real-world network datasets show that the node representations learned with MVGE achieve significant performance improvements in three different downstream tasks, especially on graphs with heterophily.
Graph representation learning (graph embedding) has led to breakthrough results in various machine learning graph-based applications such as node classification, link prediction and recommendation. Many real-world graphs can be characterized as ...
To condense different feature signals into the embeddings, this paper proposes PairE, a novel unsupervised graph embedding method using two paired nodes as the basic unit of embedding to retain the high-frequency signals between nodes to support node-related and edge-related tasks.
Graph Representation Learning Beyond Node and Homophily. PairE, a novel unsupervised graph embedding method using two paired nodes as the basic unit of embedding to retain the high-frequency signals between nodes to support node-related and edge-related tasks is proposed.
Fig. 1: Inferable relations based on pairs’ feature patterns - "Graph Representation Learning Beyond Node and Homophily"
At the core of our method is learning to (1) decompose a given graph into two components, (2) extract complementary graph signals from these two components, and (3) adaptively integrate the graph signals for node classification.
"Graph Representation Learning Beyond Node and Homophily" We provide the source code of paire. Our model implementation is based on keras, which allows the model to be trained by GPU.
Graph Representation Learning Beyond Node and Homophily You Li, Bei Lin, Binli Luo, Ning Gui* Member, IEEE, Abstract—Unsupervised graph representation learning aims to distill various graph information into a downstream task-agnostic dense vector embedding.
We investigate the representation power of graph neural networks in the semi-supervised node classification task under heterophily or low homophily, i.e., in networks where connected nodes may have different class labels and dissimilar features.