DSLR：面向基于排练的图持续学习的多样性增强与结构学习
DSLR: Diversity Enhancement and Structure Learning for Rehearsal-based Graph Continual Learning
AI总结：
针对基于排练的图持续学习中重放节点代表性单一、易集中过拟合并受无关邻居影响的问题，本文提出DSLR，通过基于覆盖的多样性选择与图结构学习提升性能和效率。
摘要：
我们研究了图持续学习（GCL）方法中基于排练的方法所使用的重放缓冲区。现有的基于排练的GCL方法会为每个类别选择最具代表性的节点，并将其存储在重放缓冲区中，以便后续训练其他任务时使用。然而，我们发现，仅考虑每个重放节点的类别代表性，会使重放节点集中在各类别中心附近，从而带来对这些区域内节点过拟合的潜在风险，并加剧灾难性遗忘。此外，由于基于排练的方法严重依赖少数重放节点来保留从先前任务中获得的知识，因此在模型训练中引入具有无关邻居的重放节点，可能会对模型性能产生显著的不利影响。本文提出了一个名为DSLR的GCL模型；具体而言，我们设计了一种基于覆盖的多样性（CD）方法，以同时考虑重放节点在每个类别内的类别代表性和多样性。此外，我们采用图结构学习（GSL）来确保重放节点连接到真正具有信息量的邻居。大量实验结果证明了DSLR的有效性和效率。我们的源代码可在 https://github.com/seungyoon-Choi/DSLR_official 获取。
英文摘要：
We investigate the replay buffer in rehearsal-based approaches for graph continual learning (GCL) methods. Existing rehearsal-based GCL methods select the most representative nodes for each class and store them in a replay buffer for later use in training subsequent tasks. However, we discovered that considering only the class representativeness of each replayed node makes the replayed nodes to be concentrated around the center of each class, incurring a potential risk of overfitting to nodes residing in those regions, which aggravates catastrophic forgetting. Moreover, as the rehearsal-based approach heavily relies on a few replayed nodes to retain knowledge obtained from previous tasks, involving the replayed nodes that have irrelevant neighbors in the model training may have a significant detrimental impact on model performance. In this paper, we propose a GCL model named DSLR, specifically, we devise a coverage-based diversity (CD) approach to consider both the class representativeness and the diversity within each class of the replayed nodes. Moreover, we adopt graph structure learning (GSL) to ensure that the replayed nodes are connected to truly informative neighbors. Extensive experimental results demonstrate the effectiveness and efficiency of DSLR. Our source code is available at https://github.com/seungyoon-Choi/DSLR_official.
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。
