Robust OT-Guided Generative Residual Domain Adaptation for Bike-Sharing Demand Prediction under Temporal Domain Shift
arXiv:2605.23115v1 Announce Type: new Abstract: Bike-sharing models trained on historical station-hour data may degrade when deployed in later years because travel patterns change over time. This paper studies March Citi Bike demand prediction from 2021 to 2026 as a temporal domain adaptation problem and proposes Gen-ROTDA, a robust optimal transport-guided residual domain adaptation framework. The method fits a target-domain station-time anchor with a small labeled target subset, transfers residual rather than raw demand, applies a deterministic label-preserving residual feature generator, an
The paper addresses a critical challenge in AI model deployment, as real-world data distributions frequently shift over time, necessitating robust adaptation techniques.
This research provides a methodological advancement for AI models operating in dynamic environments, improving reliability and reducing retraining costs for predictive systems.
Better methods for handling temporal domain shift in predictive AI models will lead to more resilient and accurate forecasting, particularly in urban planning and logistics.
- · AI/ML researchers
- · Urban planning departments
- · Logistics and transportation companies
- · Companies relying on static predictive models
- · Inefficient AI deployment strategies
Improved accuracy and stability of predictive AI models in dynamic environments such as transportation networks.
Reduced operational costs and increased efficiency for services reliant on time-sensitive forecasting.
Enhanced trust and broader adoption of AI solutions for real-world problems with evolving data patterns.
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