arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2601.02451cs.LGcs.AI
收藏

mHC-GNN:用于图神经网络的流形约束超连接

mHC-GNN: Manifold-Constrained Hyper-Connections for Graph Neural Networks

National Institute of Science Education and Research(国家科学教育与研究 institute)

Subhankar Mishra

AI总结:

研究针对图神经网络的过度平滑及表达能力受限问题,提出mHC-GNN方法,通过流形约束超连接扩展节点表示,经双机制分析减轻过度平滑,能区分超越1-WL的图,实验验证其在深度网络中效果显著,消融实验凸显流形约束的重要性。

摘要:

图神经网络(GNNs)在深度架构中存在过度平滑问题,且表达能力受限于1-维斯费勒-莱曼(1-WL)测试。我们将最近为Transformer提出的流形约束超连接应用于图神经网络。我们的方法mHC-GNN在n个并行流上扩展节点表示,并通过Sinkhorn-Knopp归一化将流混合矩阵约束到双随机矩阵的伯克霍夫多面体。我们证明mHC-GNN通过层残差下界减轻过度平滑,表明节点对差异以(1-ε)^L的速率衰减(其中ε衡量混合矩阵与单位矩阵的偏差),远慢于由谱隙γ驱动的标准(1-γ)^L坍缩速率。这种双机制分析通过L<n时的受保护正交子空间和L≥n时的层收缩,提供了先前方法所没有的与架构无关的速率保证。通过独立随机流初始化,mHC-GNN可以通过双随机混合在各层保持流多样性来区分超越1-WL的图。深度实验表明,标准GNN在超过16层后坍缩到接近随机的性能,而mHC-GNN在128层时保持超过74%的准确率,在极深时提高超过50个百分点。消融实验证实流形约束至关重要:去除它们会导致高达82%的性能下降。在异质图(罗马帝国、宾夕法尼亚94、天才)和表达能力基准(EXP)上的实验进一步验证了这一贡献。代码可在这个https URL获取。

英文摘要:

Graph Neural Networks (GNNs) suffer from over-smoothing in deep architectures and expressiveness bounded by the 1-Weisfeiler-Leman (1-WL) test. We adapt Manifold-Constrained Hyper-Connections, recently proposed for Transformers, to graph neural networks. Our method, \mhcgnn{}, expands node representations across $n$ parallel streams and constrains stream-mixing matrices to the Birkhoff polytope of doubly stochastic matrices via Sinkhorn-Knopp normalization. We prove that \mhcgnn{} mitigates over-smoothing via a layer-wise residual lower bound showing that node-pair differences decay at rate $(1{-}\varepsilon)^L$ (where $\varepsilon$ measures deviation of the mixing matrix from identity), far slower than the standard $(1{-}γ)^L$ collapse rate driven by the spectral gap $γ$. This two-regime analysis, via the protected orthogonal subspace for $L < n$ and the layer-wise contraction for $L \geq n$, provides architecture-agnostic rate guarantees absent from prior methods. With independent random stream initialization, \mhcgnn{} can distinguish graphs beyond 1-WL by maintaining stream diversity across layers via doubly stochastic mixing. Depth experiments spanning 2 to 128 layers reveal that standard GNNs collapse to near-random performance beyond 16 layers, while \mhcgnn{} maintains over 74\% accuracy at 128 layers, with improvements exceeding 50 percentage points at extreme depths. Ablations confirm that manifold constraints are essential: removing them causes up to 82\% performance degradation. Experiments on heterophilic graphs (roman-empire, penn94, genius) and expressiveness benchmarks (EXP) further validate the contribution. Code is available at https://github.com/smishra-lab/mhc-gnn

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。

↑