Image de couverture de livre

Residual Connections and Normalization Can Provably Prevent Oversmoothing in GNNs

Fuente: arXiv

Enregistré dans:

Détails bibliographiques
Auteurs principaux:	Scholkemper, Michael, Wu, Xinyi, Jadbabaie, Ali, Schaub, Michael T.
Format:	Preprint
Publié:	2024
Sujets:	Machine Learning
Accès en ligne:	Acceder al recurso
Tags:	Ajouter un tag Pas de tags, Soyez le premier à ajouter un tag!

Documents similaires

Demystifying Oversmoothing in Attention-Based Graph Neural Networks
par: Wu, Xinyi, et autres
Publié: (2023)

A Signed Graph Approach to Understanding and Mitigating Oversmoothing in GNNs
par: Wang, Jiaqi, et autres
Publié: (2025)

Graph Neural Networks Do Not Always Oversmooth
par: Epping, Bastian, et autres
Publié: (2024)

Adaptive Initial Residual Connections for GNNs with Theoretical Guarantees
par: Shirzadi, Mohammad, et autres
Publié: (2025)

Causal Cellular Context Transfer Learning (C3TL): An Efficient Architecture for Prediction of Unseen Perturbation Effects
par: Scholkemper, Michael, et autres
Publié: (2026)

Mitigating Oversmoothing Through Reverse Process of GNNs for Heterophilic Graphs
par: Park, MoonJeong, et autres
Publié: (2024)

Understanding Oversmoothing in Diffusion-Based GNNs From the Perspective of Operator Semigroup Theory
par: Zhao, Weichen, et autres
Publié: (2024)

On the Emergence of Position Bias in Transformers
par: Wu, Xinyi, et autres
Publié: (2025)

Solving Oversmoothing in GNNs via Nonlocal Message Passing: Algebraic Smoothing and Depth Scalability
par: Guan, Weiqi, et autres
Publié: (2025)

Setting the Record Straight on Transformer Oversmoothing
par: Dovonon, Gbètondji J-S, et autres
Publié: (2024)

On the Role of Attention Masks and LayerNorm in Transformers
par: Wu, Xinyi, et autres
Publié: (2024)

Resolving Oversmoothing with Opinion Dissensus
par: Wang, Keqin, et autres
Publié: (2025)

LMC: Fast Training of GNNs via Subgraph Sampling with Provable Convergence
par: Shi, Zhihao, et autres
Publié: (2023)

A least-square method for non-asymptotic identification in linear switching control
par: Sun, Haoyuan, et autres
Publié: (2024)

Online Learning for Supervisory Switching Control
par: Sun, Haoyuan, et autres
Publié: (2026)

Position: Message-passing and spectral GNNs are two sides of the same coin
par: Vasileiou, Antonis, et autres
Publié: (2026)

Distance-Restricted Folklore Weisfeiler-Leman GNNs with Provable Cycle Counting Power
par: Zhou, Junru, et autres
Publié: (2023)

Oversmoothing: A Nightmare for Graph Contrastive Learning?
par: Li, Jintang, et autres
Publié: (2023)

CORGI: GNNs with Convolutional Residual Global Interactions for Lagrangian Simulation
par: Ji, Ethan, et autres
Publié: (2025)

Overcoming Oversmoothness in Graph Convolutional Networks via Hybrid Scattering Networks
par: Wenkel, Frederik, et autres
Publié: (2022)

Looped Transformers with Layer Normalization Provably Learn the Power Method
par: Wu, Lyumin, et autres
Publié: (2026)

Fixed Aggregation Features Can Rival GNNs
par: Rubio-Madrigal, Celia, et autres
Publié: (2026)

Extrapolation Guarantees for Perturbation Modeling Under the Additive Latent Shift Assumption
par: von Kügelgen, Julius, et autres
Publié: (2025)

Learning From Simplicial Data Based on Random Walks and 1D Convolutions
par: Frantzen, Florian, et autres
Publié: (2024)

The Pitfalls of Imitation Learning when Actions are Continuous
par: Simchowitz, Max, et autres
Publié: (2025)

A Test-Function Approach to Incremental Stability
par: Pfrommer, Daniel, et autres
Publié: (2025)

On the Role of Transformer Feed-Forward Layers in Nonlinear In-Context Learning
par: Sun, Haoyuan, et autres
Publié: (2025)

Taming Gradient Oversmoothing and Expansion in Graph Neural Networks
par: Park, MoonJeong, et autres
Publié: (2024)

Partially Trained Graph Convolutional Networks Resist Oversmoothing
par: Kelesis, Dimitrios, et autres
Publié: (2024)

The Oversmoothing Fallacy: A Misguided Narrative in GNN Research
par: Park, MoonJeong, et autres
Publié: (2025)

Global Ground Metric Learning with Applications to scRNA data
par: Kühn, Damin, et autres
Publié: (2025)

Classic GNNs are Strong Baselines: Reassessing GNNs for Node Classification
par: Luo, Yuankai, et autres
Publié: (2024)

Only Large Weights (And Not Skip Connections) Can Prevent the Perils of Rank Collapse
par: Alman, Josh, et autres
Publié: (2025)

Understanding GNNs and Homophily in Dynamic Node Classification
par: Ito, Michael, et autres
Publié: (2025)

How to escape sharp minima with random perturbations
par: Ahn, Kwangjun, et autres
Publié: (2023)

Federated Optimization of Smooth Loss Functions
par: Jadbabaie, Ali, et autres
Publié: (2022)

Replay Can Provably Increase Forgetting
par: Mahdaviyeh, Yasaman, et autres
Publié: (2025)

Can Classic GNNs Be Strong Baselines for Graph-level Tasks? Simple Architectures Meet Excellence
par: Luo, Yuankai, et autres
Publié: (2025)

Locality-Aware Graph-Rewiring in GNNs
par: Barbero, Federico, et autres
Publié: (2023)

Are We Measuring Oversmoothing in Graph Neural Networks Correctly?
par: Zhang, Kaicheng, et autres
Publié: (2025)