受人脑启发的双记忆机制，让车辆轨迹预测模型持续学习不遗忘。
Complementary Learning System Empowers Online Continual Learning of Vehicle Motion Forecasting in Smart Cities
- 模仿人脑双记忆系统，动态协调长期与短期知识。
- 在超77万辆车数据上，遗忘率降低74.31%，计算量减少94.02%。
- 无需额外数据，适合智慧城市中实时持续学习场景。
人工智能支撑多数智慧城市场景，但基于深度神经网络（DNN）的车辆轨迹预测仍面临灾难性遗忘问题——模型更新时会丢失早期知识。传统方法通过扩大训练集或重放历史数据修复，但成本高、采样效率低，且难以平衡长短期经验，无法实现类人持续学习。本文提出Dual-LS，一种无任务依赖的在线持续学习范式，受人脑互补学习系统启发。该方法结合两种协同的记忆重放机制，加速经验检索并动态协调长期与短期知识表征。在覆盖三个国家、超过77.2万辆车、累计测试里程达11,187公里的真实数据上验证，Dual-LS可将灾难性遗忘降低74.31%，计算资源需求减少94.02%，显著提升预测稳定性，且无需增加数据量。同时赋予DNN驱动的车辆轨迹预测系统高效、类人持续学习能力，适用于智慧城市的实际部署。
原文摘要 · Abstract (English)
Artificial intelligence underpins most smart city services, yet deep neural network (DNN) that forecasts vehicle motion still struggle with catastrophic forgetting, the loss of earlier knowledge when models are updated. Conventional fixes enlarge the training set or replay past data, but these strategies incur high data collection costs, sample inefficiently and fail to balance long- and short-term experience, leaving them short of human-like continual learning. Here we introduce Dual-LS, a task-free, online continual learning paradigm for DNN-based motion forecasting that is inspired by the complementary learning system of the human brain. Dual-LS pairs two synergistic memory rehearsal replay mechanisms to accelerate experience retrieval while dynamically coordinating long-term and short-term knowledge representations. Tests on naturalistic data spanning three countries, over 772,000 vehicles and cumulative testing mileage of 11,187 km show that Dual-LS mitigates catastrophic forgetting by up to 74.31\% and reduces computational resource demand by up to 94.02\%, markedly boosting predictive stability in vehicle motion forecasting without inflating data requirements. Meanwhile, it endows DNN-based vehicle motion forecasting with computation efficient and human-like continual learning adaptability fit for smart cities.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。