arXiv：cs.LG· Mei Wu, Wenchao Weng, Wenxin Su, Wenjie Tang, Wei Zhou·· 5 小时前AI 评分33
CoMemNet：面向交通预测的漂移感知采样持续记忆网络
CoMemNet: A Continual Memory Network with Drift-Aware Sampling for Traffic Prediction
AI 导读
CoMemNet 是一种面向演化交通传感器网络的持续记忆网络，通过 Online 分支适应当前时段、EMA Target 分支提供稳定特征参考，并用基于 Wasserstein 的 Drift Sampler 筛选少量漂移敏感节点更新。
正文
Abstract:Traffic sensor networks evolve as sensors are added and traffic distributions change, whereas most forecasting models assume a fixed node set and repeatedly retrain on all available data. We propose CoMemNet, a Continual Memory Network for efficient prediction over evolving traffic sensor networks. CoMemNet uses an Online branch to adapt to the current period and an exponential-moving-average Target branch as a stable feature reference. A Wasserstein-based Drift Sampler compares node-wise Online-Target feature distributions and selects a limited set of drift-sensitive nodes for updating. A lightweight Node-Adaptive Temporal Memory Replay Buffer (TMRB-N) retains compact temporal states without repeatedly traversing all historical training data. The prediction backbone does not consume an adjacency matrix; sensor adjacency is used only to construct data and optionally expand the selected update set to a limited neighborhood. Experiments on three multi-period PeMS datasets include three-seed evaluation, strong static retraining and continual baselines, controlled sampling strategies, continual-learning metrics, robustness tests, and resource accounting. The results show that CoMemNet maintains stable prediction accuracy and efficient adaptation under bounded shared-node selection, achieving a better balance between historical knowledge preservation and current-period prediction performance. Meanwhile, as the evolving network expands, CoMemNet shows clearer accuracy and cumulative training-time advantages over current-period retraining baselines. The code is available at:this https URL.
| Comments: | Accepted by IEEE Transactions on Computational Social Systems (TCSS) |
| Subjects: | Machine Learning (cs.LG); Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2605.05738 [cs.LG] |
| (or arXiv:2605.05738v2 [cs.LG] for this version) | |
| https://doi.org/10.48550/arXiv.2605.05738 arXiv-issued DOI via DataCite |
Submission history
From: Mei Wu [view email]
[v1]
Thu, 7 May 2026 06:29:58 UTC (5,617 KB)
[v2]
Fri, 2 Oct 2026 10:37:07 UTC (5,878 KB)
来源：arXiv：cs.LG · arxiv.org