报告题目：Hydrologic deep learning : origin, uncertainties, forecast and beyond

报告人：Chaopeng Shen副教授，美国宾夕法尼亚州立大学
邀请人：刘攀  教授

语言：中文
时间：2019年7月3日（星期三）下午2:30

地点：工学部八教8213会议室

报告人简介：

      Chaopeng Shen，博士，宾夕法尼亚州立大学土木工程系副教授。

      2009年，在密歇根州立大学取得environmental engineering的博士学位。2011-2012年，担任伯克利劳伦斯伯克利国家实验室研究助理(Post-Doctoral Research Associate)。现担任Water Resources Research、Frontiers in Artificial Intelligence杂志副主编。

      研究领域：His recent efforts focused on harnessing the big data and machine learning opportunities in advancing hydrologic predictions and understanding. He has written technical, editorial, review and collective opinion papers on hydrologic deep learning to call to attention the emerging opportunities for scientific advances. In addition, his research interests also include floodplain systems, scaling issues, process-based hydrologic modeling, and hydrologic data mining.

报告简介：

     Recently, deep learning (DL) has emerged as a revolutionary and versatile tool transforming industrial applications and scientific discovery. In this talk I discuss recent applications of hydrologic DL from both our group and the community. Using soil moisture and streamflow as examples, I demonstrate the power of long short-term memory (LSTM) for mimicking dynamical systems for the purpose of long-term projection, short-term forecast, and obtaining hydrologic insights. We evaluate uncertainty estimates with our DL models and show out data ingestion, which greatly improves forecast accuracy, can be flexibly implemented with DL. It is argued here that hydrologic DL opens up an alternative and transformative avenue toward operational predictions and hydrologic knowledge discovery. Further, I outline several key steps that the community, together, can help incubate progress in hydrologic DL, including open competitions, integration with process-based models, open models and datasets, and revamped educational programs.

      欢迎相关专业教师和研究生的光临！