USER: Unsupervised Structural Entropy-based Robust Graph Neural Network

2023-02-12 · Yifei Wang, Yupan Wang, Zeyu Zhang, Song Yang, Kaiqi Zhao, Jiamou Liu

Unsupervised/self-supervised graph neural networks (GNN) are vulnerable to inherent randomness in the input graph data which greatly affects the performance of the model in downstream tasks. In this paper, we alleviate the interference of graph randomness and learn appropriate representations of nodes without label information. To this end, we propose USER, an unsupervised robust version of graph neural networks that is based on structural entropy. We analyze the property of intrinsic connectivity and define intrinsic connectivity graph. We also identify the rank of the adjacency matrix as a crucial factor in revealing a graph that provides the same embeddings as the intrinsic connectivity graph. We then introduce structural entropy in the objective function to capture such a graph. Extensive experiments conducted on clustering and link prediction tasks under random-noises and meta-attack over three datasets show USER outperforms benchmarks and is robust to heavier randomness.

📄 PDF Abstract BibTeX arXiv:2302.05889

Code (1)

wangyifeibeijing/user 공식 구현 pytorch

Tasks

Graph Neural Network Link PredictionNode Clustering

Unsupervised Graph Clustering with Deep Structural Entropy

2025-05-20 · Jingyun Zhang, Hao Peng, Li Sun, Guanlin Wu 외

Research on Graph Structure Learning (GSL) provides key insights for graph-based clustering, yet current methods like Graph Neural Networks (GNNs), Graph Attention Networks (GATs), and contrastive learning often rely hea…

Clustering Contrastive Learning Graph Attention Graph Clustering+1

Unsupervised Skin Lesion Segmentation via Structural Entropy Minimization on Multi-Scale Superpixel Graphs

2023-09-05 · Guangjie Zeng, Hao Peng, Angsheng Li, Zhiwei Liu 외

Skin lesion segmentation is a fundamental task in dermoscopic image analysis. The complex features of pixels in the lesion region impede the lesion segmentation accuracy, and existing deep learning-based methods often la…

Lesion Segmentation Outlier Detection Segmentation Skin Lesion Segmentation

Breaking Structural Isolation: Scalable Graph Clustering via Community-Aware Sampling and Structural Entropy

2026-07-06 · Jingyun Zhang, Hao Peng, Jianxin Li, Angsheng Li 외 arxiv

Unsupervised graph clustering is a fundamental technique for uncovering underlying semantic patterns in large-scale networks. Although Graph Contrastive Learning has demonstrated promising performance, existing methods o…

Contrastive Learning Graph Clustering

Structural Entropy Guided Unsupervised Graph Out-Of-Distribution Detection

2025-03-05 · Yue Hou, He Zhu, Ruomei Liu, Yingke Su 외

With the emerging of huge amount of unlabeled data, unsupervised out-of-distribution (OOD) detection is vital for ensuring the reliability of graph neural networks (GNNs) by identifying OOD samples from in-distribution (…

Contrastive Learning Graph Classification Out-of-Distribution Detection Out of Distribution (OOD) Detection+1

MRGSEM-Sum: An Unsupervised Multi-document Summarization Framework based on Multi-Relational Graphs and Structural Entropy Minimization

2025-07-31 · Yongbing Zhang, Fang Nan, Shengxiang Gao, Yuxin Huang 외 arxiv

The core challenge faced by multi-document summarization is the complexity of relationships among documents and the presence of information redundancy. Graph clustering is an effective paradigm for addressing this issue,…

Multi-Document Summarization Graph Clustering