图终身学习:综述
Lifelong Graph Learning:A Comprehensive Review
刘壮 1董子宸 1董宜琳 1尚家名 1张帆 1陈雨然 1楼佩妍 1孙欣然 1王昱 1赵军 2Wayne Lin3
作者信息
- 1. 东北财经大学金融科技学院 辽宁大连 116025
- 2. 大连理工大学计算机学院 辽宁大连 116024
- 3. 南加州大学计算机学院 美国加利福尼亚州洛杉矶 90007
- 折叠
摘要
图终身学习(lifelong graph learning,LGL)是一个新兴领域,旨在实现对图结构数据的持续学习,以解决现有任务上的灾难性遗忘问题,并使得顺序更新的模型能够适应新出现的图任务.尽管LGL展现出良好的学习能力,但如何持续提高其性能仍然是一个至关重要的问题.为填补现有研究对这一方面的空白,对最近在LGL领域的研究进行了全面调查和总结.首先,重新分类了LGL的现有方法,重点关注克服灾难性遗忘的方法.随后,系统地分析了这些方法的优缺点,并探讨了实现持续性能提升的潜在解决方案.该研究着重于如何在持续学习的过程中避免对旧任务的遗忘,同时快速适应新任务的挑战.最后,还就LGL的未来发展方向进行了讨论,涵盖了其在应用领域、开放性问题等方面的潜在影响,并具体分析了这些方向对持续性能改进的潜在影响.这些讨论将有助于指导未来LGL研究的方向,推动这一领域的进一步发展与应用.
Abstract
Lifelong Graph Learning(LGL)is an emerging field aimed at achieving continuous learning on graph-structured data,addressing the catastrophic forgetting issue in existing tasks and enabling models to adapt to newly emerging graph tasks in a sequential manner.Despite demonstrating strong learning capabilities,improving LGL's sustained performance remains a crucial challenge.To address the gaps in existing research,we provide a comprehensive survey and summary of recent developments in the field of LGL.Firstly,we reclassify existing methods of LGL,focusing particularly on approaches to overcome catastrophic forgetting.Secondly,we systematically analyze the strengths and weaknesses of these methods,and discuss potential solutions for achieving sustained performance improvements.Our study emphasizes on how to avoid forgetting old tasks during the continual learning process while swiftly adapting to the challenges of new tasks.Finally,we also discuss the future directions of LGL,covering potential impacts in the application domains,open issues,and specifically analyze their potential effects on sustained performance improvements.These discussions will help guide future research directions in LGL,promoting further development and application in this field.
关键词
图终身学习(LGL)/图神经网络(GNN)/重放/正则化/持续学习/增量学习/灾难性遗忘Key words
lifelong graph learning/graph neural networks/replay/regularization/continual learning/incremental learning/catastrophic forgetting引用本文复制引用
出版年
2024
NETL
NSTL
万方数据