On the Necessity of Learnable Sheaf Laplacians

• URL: http://arxiv.org/abs/2603.05395v1
• Date: Thu, 05 Mar 2026 17:24:13 GMT
• Title: On the Necessity of Learnable Sheaf Laplacians
• Authors: Ferran Hernandez Caralt, Mar Gonzàlez i Català, Adrián Bazaga, Pietro Liò,
• Abstract summary: Sheaf Neural Networks (SNNs) were introduced as an extension of Graph Convolutional Networks to address oversmoothing on heterophilous graphs.<n>We show that Identity Sheaf Networks do not appear to suffer more significant oversmoothing than their SNN counterparts.
• Score: 12.949900291633943
• License: http://creativecommons.org/licenses/by-nc-nd/4.0/
• Abstract: Sheaf Neural Networks (SNNs) were introduced as an extension of Graph Convolutional Networks to address oversmoothing on heterophilous graphs by attaching a sheaf to the input graph and replacing the adjacency-based operator with a sheaf Laplacian defined by (learnable) restriction maps. Prior work motivates this design through theoretical properties of sheaf diffusion and the kernel of the sheaf Laplacian, suggesting that suitable non-identity restriction maps can avoid representations converging to constants across connected components. Since oversmoothing can also be mitigated through residual connections and normalization, we revisit a trivial sheaf construction to ask whether the additional complexity of learning restriction maps is necessary. We introduce an Identity Sheaf Network baseline, where all restriction maps are fixed to the identity, and use it to ablate the empirical improvements reported by sheaf-learning architectures. Across five popular heterophilic benchmarks, the identity baseline achieves comparable performance to a range of SNN variants. Finally, we introduce the Rayleigh quotient as a normalized measure for comparing oversmoothing across models and show that, in trained networks, the behavior predicted by the diffusion-based analysis of SNNs is not reflected empirically. In particular, Identity Sheaf Networks do not appear to suffer more significant oversmoothing than their SNN counterparts.

Related papers

• Variational Bayesian Flow Network for Graph Generation [54.94088904387278]
  We propose Variational Bayesian Flow Network (VBFN) for graph generation.<n>VBFN performs variational lifting to a tractable joint Gaussian variational belief family governed by structured precisions.<n>On synthetic and molecular graph datasets, VBFN improves fidelity and diversity, and surpasses baseline methods.
  arXiv  Detail & Related papers  (2026-01-30T03:59:38Z)
• ReDiSC: A Reparameterized Masked Diffusion Model for Scalable Node Classification with Structured Predictions [64.17845687013434]
  We propose ReDiSC, a structured diffusion model for structured node classification.<n>We show that ReDiSC achieves superior or highly competitive performance compared to state-of-the-art GNN, label propagation, and diffusion-based baselines.<n> Notably, ReDiSC scales effectively to large-scale datasets on which previous structured diffusion methods fail due to computational constraints.
  arXiv  Detail & Related papers  (2025-07-19T04:46:53Z)
• Learning local discrete features in explainable-by-design convolutional neural networks [0.0]
  We introduce an explainable-by-design convolutional neural network (CNN) based on the lateral inhibition mechanism. The model consists of the predictor, that is a high-accuracy CNN with residual or dense skip connections. By collecting observations and directly calculating probabilities, we can explain causal relationships between motifs of adjacent levels.
  arXiv  Detail & Related papers  (2024-10-31T18:39:41Z)
• Joint Diffusion Processes as an Inductive Bias in Sheaf Neural Networks [14.224234978509026]
  Sheaf Neural Networks (SNNs) naturally extend Graph Neural Networks (GNNs) We propose two novel sheaf learning approaches that provide a more intuitive understanding of the involved structure maps. In our evaluation, we show the limitations of the real-world benchmarks used so far on SNNs.
  arXiv  Detail & Related papers  (2024-07-30T07:17:46Z)
• Bundle Neural Networks for message diffusion on graphs [28.19981068815194]
  We show that Bundle Neural Networks (BuNNs) can approximate any feature transformation over nodes on any graphs given injective positional encodings.<n>We also prove that BuNNs can approximate any feature transformation over nodes on any family of graphs given injective positional encodings, resulting in universal node-level expressivity.
  arXiv  Detail & Related papers  (2024-05-24T13:28:48Z)
• SCAAT: Improving Neural Network Interpretability via Saliency Constrained Adaptive Adversarial Training [10.716021768803433]
  Saliency map is a common form of explanation illustrating the heatmap of feature attributions. We propose a model-agnostic learning method called Saliency Constrained Adaptive Adversarial Training (SCAAT) to improve the quality of such DNN interpretability.
  arXiv  Detail & Related papers  (2023-11-09T04:48:38Z)
• Geometric Graph Filters and Neural Networks: Limit Properties and Discriminability Trade-offs [122.06927400759021]
  We study the relationship between a graph neural network (GNN) and a manifold neural network (MNN) when the graph is constructed from a set of points sampled from the manifold. We prove non-asymptotic error bounds showing that convolutional filters and neural networks on these graphs converge to convolutional filters and neural networks on the continuous manifold.
  arXiv  Detail & Related papers  (2023-05-29T08:27:17Z)
• Entangled Residual Mappings [59.02488598557491]
  We introduce entangled residual mappings to generalize the structure of the residual connections. An entangled residual mapping replaces the identity skip connections with specialized entangled mappings. We show that while entangled mappings can preserve the iterative refinement of features across various deep models, they influence the representation learning process in convolutional networks.
  arXiv  Detail & Related papers  (2022-06-02T19:36:03Z)
• 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)
• Learning Chebyshev Basis in Graph Convolutional Networks for Skeleton-based Action Recognition [14.924672048447338]
  Spectral graph convolutional networks (GCNs) are particular deep models which aim at extending neural networks to arbitrary irregular domains. We introduce a novel spectral GCN that learns not only the usual convolutional parameters but also the Laplacian operators.
  arXiv  Detail & Related papers  (2021-04-12T14:08:58Z)
• Discriminability of Single-Layer Graph Neural Networks [172.5042368548269]
  Graph neural networks (GNNs) have exhibited promising performance on a wide range of problems. We focus on the property of discriminability and establish conditions under which the inclusion of pointwise nonlinearities to a stable graph filter bank leads to an increased discriminative capacity for high-eigenvalue content.
  arXiv  Detail & Related papers  (2020-10-17T18:52:34Z)

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.