arXiv：cs.LG· Md Abrar Jahin, Taufikur Rahman Fuad, Md Rizwan Parvez·· 3 小时前AI 评分31
KAIROS：面向动态图对比学习的可微 Koopman 算子
Differentiable Koopman Operator for Contrastive Learning on Dynamic Graphs
AI 导读
研究者提出 KAIROS，一个将可微 Koopman 算子嵌入动态图对比学习循环的自监督框架，用于在学习到的嵌入空间中对时间演化进行线性化。该框架采用双视图编码器，将原始节点特征与图扩散结构视图配对，并在多个时间窗口上以多粒度对比目标优化。在九个动态图基准上，KAIROS 全部取得 SOTA 异常检测结果，较此前工作最高提升 23.15 ROC-AUC 点，同时在无监督节点分类上保持竞争力。
正文
Abstract:Real-world interaction networks are inherently dynamic: edges form and dissolve as node behavior shifts over time. Most snapshot-based contrastive methods encode temporal dependencies implicitly in encoder weights, without an explicit model of how node representations evolve, making them brittle under distribution shifts. We propose KAIROS (Koopman-Aligned Invariant Representations for Open Dynamic Systems), a self-supervised framework that embeds a differentiable Koopman operator within a dynamic graph contrastive learning loop to linearize temporal evolution in the learned embedding space. A dual-view encoder pairs raw node features with a graph-diffused structural view and is optimized with multi-granularity contrastive objectives across temporal windows. For anomaly detection, KAIROS uses the Koopman prediction residual together with temporal inconsistency and local neighborhood deviation to separate irregular behavior from predictable graph evolution. Evaluated on nine dynamic graph benchmarks, KAIROS achieves state-of-the-art anomaly detection results on all nine datasets, with gains of up to 23.15 ROC-AUC points over prior work, while remaining competitive for unsupervised node classification. These results show that explicit dynamics modeling provides a scalable and effective inductive bias for temporal graph representation learning.
| Subjects: | Machine Learning (cs.LG) |
| Cite as: | arXiv:2610.02990 [cs.LG] |
| (or arXiv:2610.02990v1 [cs.LG] for this version) | |
| https://doi.org/10.48550/arXiv.2610.02990 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Taufikur Rahman Fuad [view email]
[v1]
Fri, 2 Oct 2026 08:24:23 UTC (221 KB)
来源：arXiv：cs.LG · arxiv.org