arXiv:2606.10499v1 Announce Type: new Abstract: Traffic prediction is fundamental to intelligent transportation systems and urban computing, yet many cities continue to suffer from traffic data scarcity due to limited sensor deployment and uneven urban development. Cross-city knowledge transfer has thus attracted increasing attention, enabling data-rich cities to assist data-scarce ones. However, centralized approaches raise privacy concerns, while existing federated methods struggle with pronounced spatiotemporal heterogeneity across cities. To address these challenges, we propose MoE-FedTP,
The increasing availability of AI research and the pressing need for efficient urban management in the face of data scarcity drive the development of advanced spatiotemporal prediction models.
This research addresses critical challenges in urban computing and intelligent transportation by providing a privacy-preserving and robust solution for cross-city data utilization, impacting infrastructure efficiency and resource allocation.
The ability to leverage data from disparate cities without centralizing sensitive information could significantly enhance urban planning, traffic management, and emergency response across diverse regions.
- · Smart city initiatives
- · Urban planners
- · Transportation authorities
- · AI/ML researchers
- · Cities with insufficient sensor infrastructure (if they fail to adopt such solut
- · Traditional, centralized data sharing models
Improved traffic prediction and urban resource management through privacy-preserving federated learning.
Reduced operational costs for transportation departments and enhanced public safety due to better predictive capabilities.
The establishment of interoperable and privacy-centric urban AI ecosystems, fostering greater cross-city collaboration on infrastructure and environmental challenges.
This signal links to a primary source. Continuum Brief monitors and indexes it as part of the live intelligence stream — we do not republish source content.
Read at arXiv cs.LG