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 2309.02012cs.LGcs.AIcs.SI
收藏

iLoRE:具有即时长期建模和重现保留的动态图表示

iLoRE: Dynamic Graph Representation with Instant Long-term Modeling and Re-occurrence Preservation

Fudan University(复旦大学);Tencent Weixin Group(腾讯微信事业群);University of California, Davis(加州大学戴维斯分校)

Siwei Zhang, Yun Xiong, Yao Zhang, Xixi Wu, Yiheng Sun, Jiawei Zhang

AI总结:

针对动态图建模中无差别更新、长期依赖建模不足和忽视重现模式的问题,提出iLoRE方法,通过自适应短期更新器、基于身份注意力的长期更新器和图模块编码重现模式,在真实数据集上验证了有效性。

摘要:

连续时间动态图建模是许多实际应用(如金融风险管理和欺诈检测)中的关键任务。尽管现有的动态图建模方法已取得令人满意的结果,但它们仍存在三个关键局限,阻碍了其可扩展性和进一步适用性。i) 无差别更新。对于传入的边,现有方法会无差别地处理它们,这可能导致更多的时间消耗和意外的噪声信息。ii) 无效的节点级长期建模。它们严重依赖循环神经网络(RNN)作为主干,而RNN已被证明无法完全捕获事件序列中的节点级长期依赖关系。iii) 忽视重现模式。动态图涉及邻居的重复出现,这指示了它们的重要性,而现有方法令人遗憾地忽视了这一点。在本文中,我们提出了iLoRE,一种新颖的动态图建模方法,具有即时节点级长期建模和重现保留。为了克服无差别更新问题,我们引入了自适应短期更新器模块,该模块将自动丢弃无用或噪声边,确保iLoRE的有效性和即时能力。我们进一步提出了长期更新器,以实现更有效的节点级长期建模,其中我们创新性地提出了身份注意力机制来赋能基于Transformer的更新器,绕过了典型RNN主导设计的有限有效性。最后,关键的重现模式也被编码到图模块中,用于信息丰富的表示学习,这将进一步提高我们方法的表达能力。我们在真实世界数据集上的实验结果证明了我们的iLoRE在动态图建模中的有效性。

英文摘要:

Continuous-time dynamic graph modeling is a crucial task for many real-world applications, such as financial risk management and fraud detection. Though existing dynamic graph modeling methods have achieved satisfactory results, they still suffer from three key limitations, hindering their scalability and further applicability. i) Indiscriminate updating. For incoming edges, existing methods would indiscriminately deal with them, which may lead to more time consumption and unexpected noisy information. ii) Ineffective node-wise long-term modeling. They heavily rely on recurrent neural networks (RNNs) as a backbone, which has been demonstrated to be incapable of fully capturing node-wise long-term dependencies in event sequences. iii) Neglect of re-occurrence patterns. Dynamic graphs involve the repeated occurrence of neighbors that indicates their importance, which is disappointedly neglected by existing methods. In this paper, we present iLoRE, a novel dynamic graph modeling method with instant node-wise Long-term modeling and Re-occurrence preservation. To overcome the indiscriminate updating issue, we introduce the Adaptive Short-term Updater module that will automatically discard the useless or noisy edges, ensuring iLoRE's effectiveness and instant ability. We further propose the Long-term Updater to realize more effective node-wise long-term modeling, where we innovatively propose the Identity Attention mechanism to empower a Transformer-based updater, bypassing the limited effectiveness of typical RNN-dominated designs. Finally, the crucial re-occurrence patterns are also encoded into a graph module for informative representation learning, which will further improve the expressiveness of our method. Our experimental results on real-world datasets demonstrate the effectiveness of our iLoRE for dynamic graph modeling.

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。

↑