arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2402.05396cs.LGcs.AI
收藏

TASER:面向快速且准确的动态图表示学习的时间自适应采样

TASER: Temporal Adaptive Sampling for Fast and Accurate Dynamic Graph Representation Learning

University of Southern California(南加州大学);Meta AI;US Army Research Lab(美国陆军研究实验室)

Gangda Deng, Hongkuan Zhou, Hanqing Zeng, Yinglong Xia, Christopher Leung, Jianbo Li, Rajgopal Kannan, Viktor Prasanna

AI总结:

本文提出TASER,首个面向时序图神经网络的自适应采样方法,通过基于训练动态的mini-batch选择和时间邻居采样,在五个数据集上相比基线平均提升2.3% MRR并实现5.1倍训练加速。

摘要:

近年来,时序图神经网络(TGNNs)在欺诈检测和内容推荐等多种高影响力应用中展现出最先进的性能。尽管TGNNs取得了成功,但它们容易受到现实世界动态图中普遍存在的噪声影响,例如时间上已失效的链接和倾斜的交互分布。这些噪声导致两个严重损害TGNN准确性的关键问题:(1)模型由劣质交互监督,(2)噪声输入导致聚合消息中的高方差。然而,当前的TGNN去噪技术并未考虑每个节点多样且动态的噪声模式。此外,它们还因遍历更多邻居而产生过高的mini-batch生成开销。我们认为,实现快速且准确的TGNNs的补救措施在于时间自适应采样。在这项工作中,我们提出了TASER,这是首个针对TGNNs的自适应采样方法,在准确性、效率和可扩展性方面进行了优化。TASER根据训练动态调整其mini-batch选择,并根据过去交互的上下文、结构和时间属性调整时间邻居选择。为了缓解mini-batch生成的瓶颈,TASER实现了一个纯基于GPU的时间邻居查找器和一个专用的GPU特征缓存。我们使用两种最先进的骨干TGNNs评估了TASER的性能。在五个流行数据集上,TASER在平均倒数排名(MRR)上平均比相应基线高出2.3%,同时在训练时间上实现了平均5.1倍的加速。

英文摘要:

Recently, Temporal Graph Neural Networks (TGNNs) have demonstrated state-of-the-art performance in various high-impact applications, including fraud detection and content recommendation. Despite the success of TGNNs, they are prone to the prevalent noise found in real-world dynamic graphs like time-deprecated links and skewed interaction distribution. The noise causes two critical issues that significantly compromise the accuracy of TGNNs: (1) models are supervised by inferior interactions, and (2) noisy input induces high variance in the aggregated messages. However, current TGNN denoising techniques do not consider the diverse and dynamic noise pattern of each node. In addition, they also suffer from the excessive mini-batch generation overheads caused by traversing more neighbors. We believe the remedy for fast and accurate TGNNs lies in temporal adaptive sampling. In this work, we propose TASER, the first adaptive sampling method for TGNNs optimized for accuracy, efficiency, and scalability. TASER adapts its mini-batch selection based on training dynamics and temporal neighbor selection based on the contextual, structural, and temporal properties of past interactions. To alleviate the bottleneck in mini-batch generation, TASER implements a pure GPU-based temporal neighbor finder and a dedicated GPU feature cache. We evaluate the performance of TASER using two state-of-the-art backbone TGNNs. On five popular datasets, TASER outperforms the corresponding baselines by an average of 2.3% in Mean Reciprocal Rank (MRR) while achieving an average of 5.1x speedup in training time.

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。

↑