随机权重让图神经网络更高效，性能接近训练好的模型。
On the Effectiveness of Random Weights in Graph Neural Networks
- 用随机权重替代可学习参数，简化训练过程。
- 训练速度提升6倍，内存占用减少3倍，性能相当。
- 缓解特征秩坍塌问题，适合追求效率的GNN应用。
图神经网络（GNN）在图结构数据上表现优异，主要依赖消息传递层中的可学习权重。本文表明，使用随机权重也能取得与端到端训练相当的性能，适用于多种任务和数据集。通过将可学习权重替换为随机权重，GNN能保持强大预测能力，同时将训练时间缩短最多6倍，内存消耗降低最多3倍。此外，结合本文提出的构造方式，随机权重生成了随机图传播算子，有效缓解了GNN中的特征秩坍塌问题。这些发现揭示了随机权重作为轻量高效替代方案的潜力，为GNN架构的设计与训练提供了新视角。
原文摘要 · Abstract (English)
Graph Neural Networks (GNNs) have achieved remarkable success across diverse tasks on graph-structured data, primarily through the use of learned weights in message passing layers. In this paper, we demonstrate that random weights can be surprisingly effective, achieving performance comparable to end-to-end training counterparts, across various tasks and datasets. Specifically, we show that by replacing learnable weights with random weights, GNNs can retain strong predictive power, while significantly reducing training time by up to 6$\times$ and memory usage by up to 3$\times$. Moreover, the random weights combined with our construction yield random graph propagation operators, which we show to reduce the problem of feature rank collapse in GNNs. These understandings and empirical results highlight random weights as a lightweight and efficient alternative, offering a compelling perspective on the design and training of GNN architectures.
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。