Urban prediction tasks refer to predicting urban indicators (, traffic, temperature, etc.) using urban big data, which is crucial for understanding the urban patterns, and further benefits the urban public administration. An empirical study indicates that there are correlated patterns among urban prediction tasks from various domains, which suggests the existence of domain-sharing knowledge. Aggregating such domain-sharing knowledge would significantly benefit urban prediction tasks. However, as a widely used learning paradigm for knowledge aggregation, existing meta-learning methods, especially gradient-based methods, can only work for single-domain tasks. To solve the problem, we propose Cross-Domain Meta-Learning (CDML), a flexible framework for aggregating domain-sharing knowledge from cross-domain urban prediction tasks. Specifically, the core architecture of CDML is the model fusion block that includes (1) meta-model, shared by cross-domain tasks for capturing domain-sharing knowledge; (2) domain-specific model, shared only by the same-domain tasks for preserving domain-specific knowledge; and (3) knowledge fusion unit, for combining both the domain-sharing/specific knowledge for good generalization. Moreover, we develop asynchronous meta-training and adaption strategy strategies to further guarantee cross-domain generalization. The extensive experimental results validate the effectiveness of the proposed framework with the superior ability of boosting existing urban prediction models, quick adaption, and the potential for simplifying models.

中文翻译：

---

通过元学习利用领域共享知识促进城市预测任务

城市预测任务是指利用城市大数据预测城市指标（交通、温度等），这对于理解城市格局至关重要，并进一步有利于城市公共管理。实证研究表明，不同领域的城市预测任务之间存在相关模式，这表明领域共享知识的存在。聚合此类领域共享知识将极大有利于城市预测任务。然而，作为一种广泛使用的知识聚合学习范式，现有的元学习方法，尤其是基于梯度的方法，只能适用于单域任务。为了解决这个问题，我们提出了跨域元学习（CDML），这是一种灵活的框架，用于聚合来自跨域城市预测任务的域共享知识。具体来说，CDML的核心架构是模型融合块，包括（1）元模型，由跨域任务共享，用于捕获域共享知识； (2) 特定领域模型，仅由相同领域的任务共享，以保存特定领域的知识； (3)知识融合单元，用于结合领域共享/特定知识以实现良好的泛化。此外，我们开发了异步元训练和适应策略，以进一步保证跨领域泛化。大量的实验结果验证了所提出的框架的有效性，具有增强现有城市预测模型的卓越能力、快速适应能力以及简化模型的潜力。