通过解耦节点表示与特征，提升小数据下时序图学习性能
A Transfer Framework for Enhancing Temporal Graph Learning in Data-Scarce Settings
- 设计双分图编码机制，将节点表示与其特征解耦
- 在低数据场景下，性能比基线高56%，优于现有迁移方法36%
- 适合数据稀缺的社交、金融等时序图预测任务
动态实体交互广泛存在于社交平台、金融系统、医疗和电子商务等领域。这些交互可建模为随时间演化的图结构，预测未来连接是推荐系统等应用的关键任务。时序图神经网络（TGNNs）在该任务上表现优异，但通常需要大量训练数据，而现实场景中数据常受限。一种缓解数据稀缺的方法是利用相关数据集上的预训练模型。然而，由于TGNN依赖节点特有记忆结构，跨数据集知识迁移困难。为此，本文提出一种新迁移框架，通过结构化双分图编码机制，将节点表示与其特征解耦，实现记忆组件及其他归纳模式的有效迁移。在真实世界基准上的实证评估表明，该方法显著提升了低数据环境下的TGNN性能，相比非迁移基线最高提升56%，优于现有迁移策略36%。
原文摘要 · Abstract (English)
Dynamic interactions between entities are prevalent in domains like social platforms, financial systems, healthcare, and e-commerce. These interactions can be effectively represented as time-evolving graphs, where predicting future connections is a key task in applications such as recommendation systems. Temporal Graph Neural Networks (TGNNs) have achieved strong results for such predictive tasks but typically require extensive training data, which is often limited in real-world scenarios. One approach to mitigating data scarcity is leveraging pre-trained models from related datasets. However, direct knowledge transfer between TGNNs is challenging due to their reliance on node-specific memory structures, making them inherently difficult to adapt across datasets. To address this, we introduce a novel transfer approach that disentangles node representations from their associated features through a structured bipartite encoding mechanism. This decoupling enables more effective transfer of memory components and other learned inductive patterns from one dataset to another. Empirical evaluations on real-world benchmarks demonstrate that our method significantly enhances TGNN performance in low-data regimes, outperforming non-transfer baselines by up to 56\% and surpassing existing transfer strategies by 36\%
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。