Deep Neural Sheaf Diffusion

• URL: http://arxiv.org/abs/2605.19021v2
• Date: Fri, 22 May 2026 18:44:04 GMT
• Title: Deep Neural Sheaf Diffusion
• Abstract summary: We propose emphDeep Neural Sheaf Diffusion (DNSD), which replaces the sheaf Laplacian with a sheaf adjacency operator to maintain informative signals across layers.<n>We demonstrate that DNSD effectively utilizes deep aggregation in graph tasks, outperforming GNN and NSD baselines with up to 30pp accuracy on synthetic long-range datasets.
• Score: 3.402787708517185
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Deep Graph Neural Networks (GNNs) are essential for capturing complex dependencies in graph-structured data. However, scaling GNNs to depth remains challenging, as stacking layers leads to representation collapse and diminishing sensitivity due to repeated aggregation. While Neural Sheaf Diffusion (NSD) provides strong theoretical guarantees against such collapse, these guarantees do not translate to practice: as depth increases, the disagreement signal of the sheaf Laplacian vanishes, limiting the contribution of deeper layers. We identify mechanisms that hinder NSD effectiveness at depth and propose \emph{Deep Neural Sheaf Diffusion} (DNSD), which replaces the sheaf Laplacian with a sheaf adjacency operator to maintain informative signals across layers. This is complemented by normalization, odd nonlinearities, and gating. To provide a principled explanation of the expected performance improvement, we contrast sheaf diffusion to graph attention mechanisms, highlighting that DNSD replaces scalar attention scores with matrix-valued edge functions and normalizes node representations rather than attention scores. We demonstrate empirically that DNSD effectively utilizes deep aggregation in graph tasks, outperforming GNN and NSD baselines with up to 30pp accuracy on synthetic long-range datasets, and consistently outperforming them on real-world benchmarks. These results position sheaf-based architectures as a promising building block for graph foundation models by supporting effective deep architectures.

Related papers

• Compositional Sparsity as an Inductive Bias for Neural Architecture Design [35.06894725394093]
  We identify the structural priors that enable Deep Neural Networks to overcome the curse of dimensionality.<n>We present an interpretable pipeline in which abstraction emerges through hierarchical composition.<n>Across a broad suite of real-world datasets, HNNs consistently match or outperform dense baselines while using far fewer parameters.
  arXiv  Detail & Related papers  (2026-05-14T12:26:50Z)
• Layer Embedding Deep Fusion Graph Neural Network [25.30982656886312]
  Graph Neural Networks (GNNs) have demonstrated impressive performance in learning representations from graph-structured data.<n>We propose a novel framework named Layer Embedding Deep Fusion Graph Neural Network (LEDF-GNN)
  arXiv  Detail & Related papers  (2026-04-25T14:25:02Z)
• Early-Exit Graph Neural Networks [18.835674898033172]
  We present Early-Exit Graph Neural Networks (EEGNNs), which append confidence-aware exit heads that allow on-the-fly termination of propagation based on each node or the entire graph.<n>EEGNNs preserve robust performance as depth grows and deliver competitive accuracy on heterophilic and long-range benchmarks.
  arXiv  Detail & Related papers  (2025-05-23T16:45:14Z)
• ScaleGNN: Towards Scalable Graph Neural Networks via Adaptive High-order Neighboring Feature Fusion [73.85920403511706]
  We propose ScaleGNN, a novel framework that adaptively fuses multi-hop node features for scalable and effective graph learning.<n>We show that ScaleGNN consistently outperforms state-of-the-art GNNs in both predictive accuracy and computational efficiency.
  arXiv  Detail & Related papers  (2025-04-22T14:05:11Z)
• Statistical physics analysis of graph neural networks: Approaching optimality in the contextual stochastic block model [0.0]
  Graph neural networks (GNNs) are designed to process data associated with graphs.<n>GNNs can encounter difficulties in gathering information from nodes far apart by iterated aggregation steps.<n>We show how the architecture of the GCN has to scale with the depth to avoid oversmoothing.
  arXiv  Detail & Related papers  (2025-03-03T09:55:10Z)
• Spatiotemporal Graph Learning with Direct Volumetric Information Passing and Feature Enhancement [62.91536661584656]
  We propose a dual-module framework, Cell-embedded and Feature-enhanced Graph Neural Network (aka, CeFeGNN) for learning.<n>We embed learnable cell attributions to the common node-edge message passing process, which better captures the spatial dependency of regional features.<n>Experiments on various PDE systems and one real-world dataset demonstrate that CeFeGNN achieves superior performance compared with other baselines.
  arXiv  Detail & Related papers  (2024-09-26T16:22:08Z)
• Deep Graph Neural Networks via Posteriori-Sampling-based Node-Adaptive Residual Module [65.81781176362848]
  Graph Neural Networks (GNNs) can learn from graph-structured data through neighborhood information aggregation. As the number of layers increases, node representations become indistinguishable, which is known as over-smoothing. We propose a textbfPosterior-Sampling-based, Node-distinguish Residual module (PSNR).
  arXiv  Detail & Related papers  (2023-05-09T12:03:42Z)
• Kernel function impact on convolutional neural networks [10.98068123467568]
  We study the usage of kernel functions at the different layers in a convolutional neural network. We show how one can effectively leverage kernel functions, by introducing a more distortion aware pooling layers. We propose Kernelized Dense Layers (KDL), which replace fully-connected layers.
  arXiv  Detail & Related papers  (2023-02-20T19:57:01Z)
• Simple and Efficient Heterogeneous Graph Neural Network [55.56564522532328]
  Heterogeneous graph neural networks (HGNNs) have powerful capability to embed rich structural and semantic information of a heterogeneous graph into node representations. Existing HGNNs inherit many mechanisms from graph neural networks (GNNs) over homogeneous graphs, especially the attention mechanism and the multi-layer structure. This paper conducts an in-depth and detailed study of these mechanisms and proposes Simple and Efficient Heterogeneous Graph Neural Network (SeHGNN)
  arXiv  Detail & Related papers  (2022-07-06T10:01:46Z)
• Deep Architecture Connectivity Matters for Its Convergence: A Fine-Grained Analysis [94.64007376939735]
  We theoretically characterize the impact of connectivity patterns on the convergence of deep neural networks (DNNs) under gradient descent training. We show that by a simple filtration on "unpromising" connectivity patterns, we can trim down the number of models to evaluate.
  arXiv  Detail & Related papers  (2022-05-11T17:43:54Z)
• EIGNN: Efficient Infinite-Depth Graph Neural Networks [51.97361378423152]
  Graph neural networks (GNNs) are widely used for modelling graph-structured data in numerous applications. Motivated by this limitation, we propose a GNN model with infinite depth, which we call Efficient Infinite-Depth Graph Neural Networks (EIGNN) We show that EIGNN has a better ability to capture long-range dependencies than recent baselines, and consistently achieves state-of-the-art performance.
  arXiv  Detail & Related papers  (2022-02-22T08:16:58Z)
• Graph Backdoor [53.70971502299977]
  We present GTA, the first backdoor attack on graph neural networks (GNNs) GTA departs in significant ways: it defines triggers as specific subgraphs, including both topological structures and descriptive features. It can be instantiated for both transductive (e.g., node classification) and inductive (e.g., graph classification) tasks.
  arXiv  Detail & Related papers  (2020-06-21T19:45:30Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.