arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2511.08094cs.LG
收藏

Stuart-Landau振荡图神经网络

Stuart-Landau Oscillatory Graph Neural Network

School of Mathematics;Maxwell Institute University of Edinburgh Edinburgh UK;Institute of Applied Mathematics;Gran Sasso Science Institute L’Aquila Italy;Maxwell Institute, \ of Edinburgh, Edinburgh UK;Maxwell Institute University of Edinburgh;Gran Sasso Science Institute;Maxwell Institute, \ of Edinburgh

Kaicheng Zhang, David N. Reynolds, Piero Deidda, Francesco Tudisco

AI总结:

本文提出基于Stuart-Landau振荡器动力学的SLGNN,通过动态演变节点特征幅度,提升图神经网络在节点分类、图分类和图回归任务中的性能。

摘要:

振荡图神经网络(OGNNs)是一类新兴的物理启发型架构,旨在缓解深度GNN中的过平滑和消失梯度问题。在本工作中,我们引入了复值Stuart-Landau图神经网络(SLGNN),这是一种基于Stuart-Landau振荡器动力学的新型架构。Stuart-Landau振荡器是Hopf分岔附近极限环行为的典型模型,是同步理论的基础,并广泛应用于如神经科学中的脑部建模。与谐振振荡器和仅相位的Kuramoto模型不同,Stuart-Landau振荡器保留了幅度和相位的动力学,能够产生丰富的现象,如幅度调节和多稳态同步。所提出的SLGNN通过允许节点特征幅度根据Stuart-Landau动力学动态演变,扩展了现有的基于Kuramoto的OGNNs,其中可调的超参数(如Hopf参数和耦合强度)提供了对特征幅度和网络结构之间相互作用的额外控制。我们在节点分类、图分类和图回归任务中进行了广泛的实验,证明SLGNN在现有OGNNs之上表现更优,并建立了一种新颖、表达力强且理论基础稳固的深度振荡架构在图上的新框架。

英文摘要:

Oscillatory Graph Neural Networks (OGNNs) are an emerging class of physics-inspired architectures designed to mitigate oversmoothing and vanishing gradient problems in deep GNNs. In this work, we introduce the Complex-Valued Stuart-Landau Graph Neural Network (SLGNN), a novel architecture grounded in Stuart-Landau oscillator dynamics. Stuart-Landau oscillators are canonical models of limit-cycle behavior near Hopf bifurcations, which are fundamental to synchronization theory and are widely used in e.g. neuroscience for mesoscopic brain modeling. Unlike harmonic oscillators and phase-only Kuramoto models, Stuart-Landau oscillators retain both amplitude and phase dynamics, enabling rich phenomena such as amplitude regulation and multistable synchronization. The proposed SLGNN generalizes existing phase-centric Kuramoto-based OGNNs by allowing node feature amplitudes to evolve dynamically according to Stuart-Landau dynamics, with explicit tunable hyperparameters (such as the Hopf-parameter and the coupling strength) providing additional control over the interplay between feature amplitudes and network structure. We conduct extensive experiments across node classification, graph classification, and graph regression tasks, demonstrating that SLGNN outperforms existing OGNNs and establishes a novel, expressive, and theoretically grounded framework for deep oscillatory architectures on graphs.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑