This repo collects papers about developing deep learning methods on disassortative graphs.
Disassortative graphs refer to those with a low node homophily. In disassortative graphs, nodes with the same label could be distant from each other and nodes with distinct labels are more likely to be connected with edges.
Please feel free to submit a pull request if you want to add good papers.
- [arXiv 2021] Two Sides of the Same Coin: Heterophily and Oversmoothing in Graph Convolutional Neural Networks [Paper]
- [ICML 2021] Graph Neural Networks Inspired by Classical Iterative Algorithms [Paper][Code]
- [Workshop on Graph Learning Benchmarks, WWW 2021] New Benchmarks for Learning on Non-Homophilous Graphs [Paper][Code]
- [ICLR 2021] How to Find Your Friendly Neighborhood: Graph Attention Design with Self-Supervision [Paper][Code]
- [ICLR 2021] Adaptive Universal Generalized PageRank Graph Neural Network [Paper][Code]
- [WSDM 2021] Node Similarity Preserving Graph Convolutional Networks [Paper][Code]
- [AAAI 2021] Graph Neural Networks with Heterophily [Paper] [Code]
- [arXiv 2020] Non-Local Graph Neural Networks [paper]
- [NeurIPS 2020] Beyond Homophily in Graph Neural Networks: Current Limitations and Effective Designs [Paper][Code]
- [ICML 2020] Simple and Deep Graph Convolutional Networks [Paper][Code]
- [KDD 2020] Residual Correlation in Graph Neural Network Regression [Paper][Code]
- [ICLR 2020] Geom-GCN: Geometric Graph Convolutional Networks [Paper][Code]