arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2112.03088cs.LGcs.AI
收藏

迁移学习改进数据稀疏区域的径流预报

Transfer learning to improve streamflow forecasts in data sparse regions

University of Missouri–Columbia(密苏里大学哥伦比亚分校);Carnegie Mellon University(卡内基梅隆大学);The Nature Conservancy(大自然保护协会);IBM Research Europe(IBM欧洲研究院)

Roland Oruche, Lisa Egede, Tracy Baker, Fearghal O'Donncha

摘要:

有效的水资源管理需要关于水的可得性的信息,包括水的质量和数量、空间和时间。在本文中,我们研究了迁移学习(TL)背后的方法,通过微调和参数转移,以提高数据稀疏区域的径流预测的泛化性能。我们提出了一个标准的递归神经网络,其形式为长短期记忆(LSTM),以适应足够大的源域数据集,并将学习到的权重重新用于更小但相似的目标域数据集。我们提出了一种方法来实现时空应用程序的迁移学习方法,通过分离模型的空间和时间组件,并训练模型,以基于表示空间变异性的分类数据集进行泛化。该框架是在美国丰富的基准数据集上开发的,并在肯尼亚大自然保护协会收集的较小数据集上进行评估。LSTM模型通过我们的TL技术表现出泛化性能。从目前的实验结果表明,有效的预测技巧,预测径流响应时,知识转移和静态描述符,以提高水文模型的推广数据稀疏的地区。

英文摘要:

Effective water resource management requires information on water availability, both in terms of quality and quantity, spatially and temporally. In this paper, we study the methodology behind Transfer Learning (TL) through fine-tuning and parameter transferring for better generalization performance of streamflow prediction in data-sparse regions. We propose a standard recurrent neural network in the form of Long Short-Term Memory (LSTM) to fit on a sufficiently large source domain dataset and repurpose the learned weights to a significantly smaller, yet similar target domain datasets. We present a methodology to implement transfer learning approaches for spatiotemporal applications by separating the spatial and temporal components of the model and training the model to generalize based on categorical datasets representing spatial variability. The framework is developed on a rich benchmark dataset from the US and evaluated on a smaller dataset collected by The Nature Conservancy in Kenya. The LSTM model exhibits generalization performance through our TL technique. Results from this current experiment demonstrate the effective predictive skill of forecasting streamflow responses when knowledge transferring and static descriptors are used to improve hydrologic model generalization in data-sparse regions.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑