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 2401.16800cs.LGcs.DMcs.SYeess.SY
收藏

时序图中节点特征预测的在线算法

Online Algorithm for Node Feature Forecasting in Temporal Graphs

University of Oxford(牛津大学)

Aniq Ur Rahman, Justin P. Coon

AI总结:

本文提出在线算法mspace,用于时序图节点特征预测,捕捉时空相关性,性能优于或媲美TGNN和卡尔曼滤波,且训练样本需求低,误差和复杂度理论有界。

摘要:

本文提出了一种用于时序图中节点特征预测的在线算法mspace,该算法能够捕捉不同节点之间的空间互相关性以及节点内部的时间自相关性。该算法可用于概率性和确定性多步预测,适用于估计和生成任务。与包括时序图神经网络(TGNN)模型和经典卡尔曼滤波器在内的多种基线进行的比较评估表明,mspace的性能与最先进方法相当,在某些数据集上甚至超越了它们。重要的是,mspace在不同训练规模的多个数据集上表现出稳定的性能,这是相对于需要大量训练样本才能有效学习数据时空趋势的TGNN模型的一个显著优势。因此,在训练样本可用性有限的情况下,采用mspace是有利的。此外,我们建立了mspace多步预测误差的理论界限,并证明其随预测步数$q$线性增长,即$\nmathcal{O}(q)$。对于渐近大的节点数$n$和时间步数$T$,mspace的计算复杂度随$n$和$T$线性增长,即$\nmathcal{O}(nT)$,而其空间复杂度保持常数$\nmathcal{O}(1)$。我们将各种mspace变体的性能与十个最新的TGNN基线以及两个经典基线ARIMA和卡尔曼滤波器在十个真实世界数据集上进行了比较。此外,我们提出了一种生成合成数据集的技术,以帮助评估节点特征预测方法,该技术有潜力作为未来研究的基准。最后,我们通过分析模型参数与数据集特征,研究了不同mspace变体的可解释性,以得出模型和数据中心的见解。

英文摘要:

In this paper, we propose an online algorithm mspace for forecasting node features in temporal graphs, which captures spatial cross-correlation among different nodes as well as the temporal auto-correlation within a node. The algorithm can be used for both probabilistic and deterministic multi-step forecasting, making it applicable for estimation and generation tasks. Comparative evaluations against various baselines, including temporal graph neural network (TGNN) models and classical Kalman filters, demonstrate that mspace performs at par with the state-of-the-art and even surpasses them on some datasets. Importantly, mspace demonstrates consistent performance across datasets with varying training sizes, a notable advantage over TGNN models that require abundant training samples to effectively learn the spatiotemporal trends in the data. Therefore, employing mspace is advantageous in scenarios where the training sample availability is limited. Additionally, we establish theoretical bounds on multi-step forecasting error of mspace and show that it scales linearly with the number of forecast steps $q$ as $\mathcal{O}(q)$. For an asymptotically large number of nodes $n$, and timesteps $T$, the computational complexity of mspace grows linearly with both $n$, and $T$, i.e., $\mathcal{O}(nT)$, while its space complexity remains constant $\mathcal{O}(1)$. We compare the performance of various mspace variants against ten recent TGNN baselines and two classical baselines, ARIMA and the Kalman filter across ten real-world datasets. Additionally, we propose a technique to generate synthetic datasets to aid in evaluating node feature forecasting methods, with the potential to serve as a benchmark for future research. Lastly, we have investigate the interpretability of different mspace variants by analyzing model parameters alongside dataset characteristics to derive model and data-centric insights.

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。

↑