paper-with-me

Papers

Demystifying Oversmoothing in Attention-Based Graph Neural Networks

2023-05-25 · NeurIPS 2023 11 · Xinyi Wu, Amir Ajorlou, Zihui Wu, Ali Jadbabaie

Oversmoothing in Graph Neural Networks (GNNs) refers to the phenomenon where increasing network depth leads to homogeneous node representations. While previous work has established that Graph Convolutional Networks (GCNs) exponentially lose expressive power, it remains controversial whether the graph attention mechanism can mitigate oversmoothing. In this work, we provide a definitive answer to this question through a rigorous mathematical analysis, by viewing attention-based GNNs as nonlinear time-varying dynamical systems and incorporating tools and techniques from the theory of products of inhomogeneous matrices and the joint spectral radius. We establish that, contrary to popular belief, the graph attention mechanism cannot prevent oversmoothing and loses expressive power exponentially. The proposed framework extends the existing results on oversmoothing for symmetric GCNs to a significantly broader class of GNN models, including random walk GCNs, Graph Attention Networks (GATs) and (graph) transformers. In particular, our analysis accounts for asymmetric, state-dependent and time-varying aggregation operators and a wide range of common nonlinear activation functions, such as ReLU, LeakyReLU, GELU and SiLU.

📄 PDF Abstract BibTeX arXiv:2305.16102

Code (0)

등록된 구현이 없습니다.

Tasks

Graph Attention

Methods 이 논문이 사용한 방법론

SiLU Sigmoid Linear Units, or SiLUs, are activation functions for neural networks. The activation of the SiLU is computed by the sigmoid function multiplied by its input, or…

Similar Papers 제목 키워드 기반

Oversmoothing, "Oversquashing", Heterophily, Long-Range, and more: Demystifying Common Beliefs in Graph Machine Learning

2025-05-21 · Adrian Arnaiz-Rodriguez, Federico Errica

After a renaissance phase in which researchers revisited the message-passing paradigm through the lens of deep learning, the graph machine learning community shifted its attention towards a deeper and practical understan…

Demystifying Oversmoothing in Sheaf Neural Networks: An Index-Theoretic Criterion

2026-08-17 · Junwen Dong, Yuhan Peng, Hao Li, Huitao Feng 외 arxiv

To combat oversmoothing in Graph Convolutional Networks, Sheaf Neural Networks (SNNs) were proposed as a generalization by equipping the graph with a sheaf structure and replacing the graph Laplacian with a sheaf Laplaci…

A Dynamical Systems-Inspired Pruning Strategy for Addressing Oversmoothing in Graph Neural Networks

2024-12-10 · Biswadeep Chakraborty, Harshit Kumar, Saibal Mukhopadhyay

Oversmoothing in Graph Neural Networks (GNNs) poses a significant challenge as network depth increases, leading to homogenized node representations and a loss of expressiveness. In this work, we approach the oversmoothin…

Diversity

Centered Self-Attention Layers

2023-06-02 · Ameen Ali, Tomer Galanti, Lior Wolf

The self-attention mechanism in transformers and the message-passing mechanism in graph neural networks are repeatedly applied within deep learning architectures. We show that this application inevitably leads to oversmo…

Weakly supervised segmentation

A Non-Asymptotic Analysis of Oversmoothing in Graph Neural Networks

2022-12-21 · Xinyi Wu, Zhengdao Chen, William Wang, Ali Jadbabaie

Oversmoothing is a central challenge of building more powerful Graph Neural Networks (GNNs). While previous works have only demonstrated that oversmoothing is inevitable when the number of graph convolutions tends to inf…

DenoisingStochastic Block Model