纠正了图神经网络深度化中的过度平滑误解，指出梯度消失才是主因。
The Oversmoothing Fallacy: A Misguided Narrative in GNN Research
- 区分了聚合、变换、激活三类操作，指出误将梯度消失归因于过度平滑
- 验证跳接和归一化可有效支持深层图神经网络无性能下降
- 适合关注深度图模型设计与理论机制的研究者阅读
过度平滑被普遍视为构建深层图神经网络（GNN）的主要障碍，限制了其性能。本文提出，过度平滑的影响被严重夸大，并呼吁重新探索深层GNN架构。基于GNN的三个核心操作——聚合、线性变换和非线性激活，我们证明先前研究错误地将梯度消失归因于过度平滑，实则由变换和激活引起，而非聚合操作。这一发现挑战了过度平滑是GNN特有问题的传统认知。此外，我们展示了跳接（skip connections）和归一化等经典方法能够实现深层GNN层的有效堆叠而无性能退化。研究澄清了对过度平滑的误解，为深层GNN的潜力提供了新视角。
原文摘要 · Abstract (English)
Oversmoothing has been recognized as a main obstacle to building deep Graph Neural Networks (GNNs), limiting the performance. This position paper argues that the influence of oversmoothing has been overstated and advocates for a further exploration of deep GNN architectures. Given the three core operations of GNNs, aggregation, linear transformation, and non-linear activation, we show that prior studies have mistakenly confused oversmoothing with the vanishing gradient, caused by transformation and activation rather than aggregation. Our finding challenges prior beliefs about oversmoothing being unique to GNNs. Furthermore, we demonstrate that classical solutions such as skip connections and normalization enable the successful stacking of deep GNN layers without performance degradation. Our results clarify misconceptions about oversmoothing and shed new light on the potential of deep GNNs.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。