Exploring Edge Disentanglement for Node Classification
Explore this paper's citation graph
Summary
This work proposes three heuristics and design three corresponding pretext tasks to guide the automatic edge disentanglement, and shows that it can achieve significant performance gains.
- Type
- article
- Published
- 2022-02-23
- Cited by
- 41
- References
- 41
- Access
- Open access
- OpenAlex
- https://openalex.org/W4224316953
- Semantic Scholar
- https://api.semanticscholar.org/CorpusID:247058599
Keywords
Computer science, Heuristics, Enhanced Data Rates for GSM Evolution, Graph, Node (physics)
References
- Spectral Networks and Locally Connected Networks on Graphs
- DeepWalk: online learning of social representations
- Neural Message Passing for Quantum Chemistry
- Graph Attention Networks
- Modeling Semantics with Gated Graph Neural Networks for Knowledge Base Question Answering
- A Comprehensive Survey on Graph Neural Networks
- Heterogeneous Graph Attention Network
- Self-Supervised Visual Feature Learning With Deep Neural Networks: A Survey
- Graph Neural Networks for Social Recommendation
- Molecular Geometry Prediction using a Deep Generative Graph Neural Network
- Survey on deep learning with class imbalance
- Disentangled Graph Convolutional Networks
- How Powerful are Graph Neural Networks?
- Diffusion-Convolutional Neural Networks
- Semi-Supervised Classification with Graph Convolutional Networks
- ALBERT: A Lite BERT for Self-supervised Learning of Language Representations
- Utilizing Molecular Network Information via Graph Convolutional Neural Networks to Predict Metastatic Event in Breast Cancer
- ChebNet: Efficient and Stable Constructions of Deep Neural Networks with Rectified Power Units using Chebyshev Approximations
- Geom-GCN: Geometric Graph Convolutional Networks
- Strategies for Pre-training Graph Neural Networks
Cited by
- Semi-supervised Drifted Stream Learning with Short Lookback
- Decoupled Self-supervised Learning for Non-Homophilous Graphs
- Link Prediction on Heterophilic Graphs via Disentangled Representation Learning
- DyTed: Disentangling Temporal Invariance and Fluctuations in Dynamic Graph Representation Learning
- TopoImb: Toward Topology-level Imbalance in Learning from Graphs
- Mitigating Popularity Bias in Recommendation with Unbalanced Interactions: A Gradient Perspective
- Curriculum Graph Poisoning
- Self-Explainable Graph Neural Networks for Link Prediction
- Skill Disentanglement for Imitation Learning from Suboptimal Demonstrations
- Learning Disentangled Representations in Signed Directed Graphs without Social Assumptions
- Graph Relation Aware Continual Learning
- Faithful and Consistent Graph Neural Network Explanations with Rationale Alignment
- Adversarial Graph Disentanglement With Component-Specific Aggregation
- Contrastive Disentangled Learning on Graph for Node Classification
- Revisiting multi-view learning: A perspective of implicitly heterogeneous Graph Convolutional Network
- Perturbation-augmented Graph Convolutional Networks: A Graph Contrastive Learning architecture for effective node classification tasks
- EsaCL: An Efficient Continual Learning Algorithm
- Learning Network Representations with Disentangled Graph Auto-Encoder
- Disambiguated Node Classification with Graph Neural Networks
- Hawkes-Enhanced Spatial-Temporal Hypergraph Contrastive Learning Based on Criminal Correlations