analiza aiqbench
CoMemNet addresses the challenge of non-stationary traffic data by introducing a drift-aware sampling mechanism that selectively updates models without retraining on full historical datasets. This approach effectively balances knowledge preservation with adaptation to evolving sensor networks, offering significant computational efficiency over standard continual learning baselines. The method demonstrates robust performance in dynamic environments where traditional fixed-node forecasting models struggle.
CoMemNet: A Continual Memory Network with Drift-Aware Sampling for Traffic Prediction
arXiv:2605.05738v2 Announce Type: replace-cross 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:https://meiwu5.github.io/CoMemNet.
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.
Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
arXiv:2605.05738 [cs.LG]
(or arXiv:2605.05738v2 [cs.LG] for this version)
https://doi.org/10.48550/arXiv.2605.05738 Focus to learn more arXiv-issued DOI via DataCite
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Text integral sub licența deschisă a publicației · 303 cuvinte
Subiecte conexe
Latent reasoning architectures would likely undermine CoT, our strongest oversight tool
AI Alignment Forum · 23 sept. 2026
News
Quantum Computing Report · 25 iul. 2026
NVIDIA Video Codec SDK 13.1: Zero-Copy Transcode, AV1 B-Frames, and Frame-Accurate Seek
NVIDIA · 31 iul. 2026
Advancing AI Infrastructure for Agentic AI with NVIDIA DOCA In-Silicon Security
NVIDIA · 1 iun. 2026