Imagen de Portada

Beyond Over-smoothing: Uncovering the Trainability Challenges in Deep Graph Neural Networks

Fuente: arXiv

Guardado en:

Detalles Bibliográficos
Autores principales:	Peng, Jie, Lei, Runlin, Wei, Zhewei
Formato:	Preprint
Publicado:	2024
Materias:	Machine Learning
Acceso en línea:	Acceder al recurso
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

Ejemplares similares

Intruding with Words: Towards Understanding Graph Injection Attacks at the Text Level
por: Lei, Runlin, et al.
Publicado: (2024)

Conjugate Learning Theory: Uncovering the Mechanisms of Trainability and Generalization in Deep Neural Networks
por: Qi, Binchuan
Publicado: (2026)

Robustness in Text-Attributed Graph Learning: Insights, Trade-offs, and New Defenses
por: Lei, Runlin, et al.
Publicado: (2025)

TIDFormer: Exploiting Temporal and Interactive Dynamics Makes A Great Dynamic Graph Transformer
por: Peng, Jie, et al.
Publicado: (2025)

Large-Scale Spectral Graph Neural Networks via Laplacian Sparsification: Technical Report
por: Ding, Haipeng, et al.
Publicado: (2025)

Exploring the Potential of Large Language Models as Predictors in Dynamic Text-Attributed Graphs
por: Lei, Runlin, et al.
Publicado: (2025)

Convolutional Neural Networks on Graphs with Chebyshev Approximation, Revisited
por: He, Mingguo, et al.
Publicado: (2022)

Future Link Prediction Without Memory or Aggregation
por: Yi, Lu, et al.
Publicado: (2025)

Better Not to Propagate: Understanding Edge Uncertainty and Over-smoothing in Signed Graph Neural Networks
por: Choi, Yoonhyuk, et al.
Publicado: (2024)

Scalable and Certifiable Graph Unlearning: Overcoming the Approximation Error Barrier
por: Yi, Lu, et al.
Publicado: (2024)

Maestro: Uncovering Low-Rank Structures via Trainable Decomposition
por: Horvath, Samuel, et al.
Publicado: (2023)

How Universal Polynomial Bases Enhance Spectral Graph Neural Networks: Heterophily, Over-smoothing, and Over-squashing
por: Huang, Keke, et al.
Publicado: (2024)

Neural Network Training via Stochastic Alternating Minimization with Trainable Step Sizes
por: Yan, Chengcheng, et al.
Publicado: (2025)

EquiPocket: an E(3)-Equivariant Geometric Graph Neural Network for Ligand Binding Site Prediction
por: Zhang, Yang, et al.
Publicado: (2023)

Depth-induced NTK: Bridging Over-parameterized Neural Networks and Deep Neural Kernels
por: Tian, Yong-Ming, et al.
Publicado: (2025)

Mamba-Based Graph Convolutional Networks: Tackling Over-smoothing with Selective State Space
por: He, Xin, et al.
Publicado: (2025)

A Trainable Feature Extractor Module for Deep Neural Networks and Scanpath Classification
por: Fuhl, Wolfgang
Publicado: (2024)

Deep Network Trainability via Persistent Subspace Orthogonality
por: Massucco, Alex, et al.
Publicado: (2025)

PolyFormer: Scalable Node-wise Filters via Polynomial Graph Transformer
por: Ma, Jiahong, et al.
Publicado: (2024)

Trainable and Explainable Simplicial Map Neural Networks
por: Paluzo-Hidalgo, Eduardo, et al.
Publicado: (2023)

From Coupled Oscillators to Graph Neural Networks: Reducing Over-smoothing via a Kuramoto Model-based Approach
por: Nguyen, Tuan, et al.
Publicado: (2023)

Quantifying the Optimization and Generalization Advantages of Graph Neural Networks Over Multilayer Perceptrons
por: Huang, Wei, et al.
Publicado: (2023)

Quantum Graph Attention Networks: Trainable Quantum Encoders for Inductive Graph Learning
por: Faria, Arthur M., et al.
Publicado: (2025)

Playing the Lottery With Concave Regularizers for Sparse Trainable Neural Networks
por: Fracastoro, Giulia, et al.
Publicado: (2025)

Over-squashing in Spatiotemporal Graph Neural Networks
por: Marisca, Ivan, et al.
Publicado: (2025)

Beyond Leakage and Complexity: Towards Realistic and Efficient Information Cascade Prediction
por: Peng, Jie, et al.
Publicado: (2025)

Trainability Beyond Linearity in Variational Quantum Objectives
por: Ma, Gordon, et al.
Publicado: (2026)

Comment on "A Note on Over-Smoothing for Graph Neural Networks"
por: Hasson, Razi, et al.
Publicado: (2025)

Uncovering Critical Sets of Deep Neural Networks via Sample-Independent Critical Lifting
por: Zhang, Leyang, et al.
Publicado: (2025)

Graph Neural Networks Uncover Geometric Neural Representations in Reinforcement-Based Motor Learning
por: Nardi, Federico, et al.
Publicado: (2024)

Rank Collapse Causes Over-Smoothing and Over-Correlation in Graph Neural Networks
por: Roth, Andreas, et al.
Publicado: (2023)

Cross-Space Adaptive Filter: Integrating Graph Topology and Node Attributes for Alleviating the Over-smoothing Problem
por: Huang, Chen, et al.
Publicado: (2024)

TGB-Seq Benchmark: Challenging Temporal GNNs with Complex Sequential Dynamics
por: Yi, Lu, et al.
Publicado: (2025)

Local Virtual Nodes for Alleviating Over-Squashing in Graph Neural Networks
por: Karabulut, Tuğrul Hasan, et al.
Publicado: (2025)

Mitigating Over-Squashing in Graph Neural Networks by Spectrum-Preserving Sparsification
por: Liang, Langzhang, et al.
Publicado: (2025)

Implicit Graph Neural Diffusion Networks: Convergence, Generalization, and Over-Smoothing
por: Fu, Guoji, et al.
Publicado: (2023)

Graph ODEs and Beyond: A Comprehensive Survey on Integrating Differential Equations with Graph Neural Networks
por: Liu, Zewen, et al.
Publicado: (2025)

Weightless Neural Networks for Continuously Trainable Personalized Recommendation Systems
por: Latif, Rafayel, et al.
Publicado: (2025)

APALU: A Trainable, Adaptive Activation Function for Deep Learning Networks
por: Subramanian, Barathi, et al.
Publicado: (2024)

LSEnet: Lorentz Structural Entropy Neural Network for Deep Graph Clustering
por: Sun, Li, et al.
Publicado: (2024)