少即是多：精心设计专家比复杂集成更高效，提升交通预测精度与速度
Less is More: Strategic Expert Selection Outperforms Ensemble Complexity in Traffic Forecasting
- 构建融合道路拓扑与特征相似性的语义空间专家，优化专家选择策略
- 在METR LA和PEMS BAY上分别实现1.3%和4.1%的MAE降低，最优配置比MegaCRN快11.5%
- 证明精简专家组合优于复杂集成，适合实时交通系统部署
交通预测对智能交通系统至关重要，有助于缓解拥堵与减少排放。现有图神经网络方法虽提升了时空建模能力，但如TESTAM等专家混合框架缺乏对物理路网拓扑的显式建模，限制了其空间表达能力。本文提出TESTAM+，引入新型语义空间专家，通过混合图构造融合物理道路结构与数据驱动特征相似性。实验表明，TESTAM+在METR LA上将MAE从3.14降至3.10（↓1.3%），在PEMS BAY上从1.72降至1.65（↓4.1%）。消融研究揭示：战略专家选择优于简单集成。单个专家表现优异：自适应专家在PEMS BAY上达1.63 MAE，优于原三专家版本（1.72）；语义空间专家同样达到1.63。最优身份+自适应配置相较SOTA MegaCRN在METR LA上降低11.5% MAE（2.99 vs. 3.38），推理延迟比完整四专家版本降低53.1%。结果表明，更少、更精准的专家设计优于复杂集成，实现新最优性能与更高计算效率，适用于实时部署。
原文摘要 · Abstract (English)
Traffic forecasting is fundamental to intelligent transportation systems, enabling congestion mitigation and emission reduction in increasingly complex urban environments. While recent graph neural network approaches have advanced spatial temporal modeling, existing mixture of experts frameworks like Time Enhanced Spatio Temporal Attention Model (TESTAM) lack explicit incorporation of physical road network topology, limiting their spatial capabilities. We present TESTAM+, an enhanced spatio temporal forecasting framework that introduces a novel SpatioSemantic Expert integrating physical road topology with data driven feature similarity through hybrid graph construction. TESTAM+ achieves significant improvements over TESTAM: 1.3% MAE reduction on METR LA (3.10 vs. 3.14) and 4.1% improvement on PEMS BAY (1.65 vs. 1.72). Through comprehensive ablation studies, we discover that strategic expert selection fundamentally outperforms naive ensemble aggregation. Individual experts demonstrate remarkable effectiveness: the Adaptive Expert achieves 1.63 MAE on PEMS BAY, outperforming the original three expert TESTAM (1.72 MAE), while the SpatioSemantic Expert matches this performance with identical 1.63 MAE. The optimal Identity + Adaptive configuration achieves an 11.5% MAE reduction compared to state of the art MegaCRN on METR LA (2.99 vs. 3.38), while reducing inference latency by 53.1% compared to the full four expert TESTAM+. Our findings reveal that fewer, strategically designed experts outperform complex multi expert ensembles, establishing new state of the art performance with superior computational efficiency for real time deployment.
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。