Please join us for a special guest seminar by Dr. Chaopeng Shen, Professor in Civil Engineering at The Pennsylvania State University, in the WRF Data Science Studio Seminar Room or online via zoom, https://washington.zoom.us/j/97451365866.

“Differentiable modeling and the Genes of AI for Water and Global sustainability” 

Abstract. Big data and artificial intelligence (AI) methods are revolutionizing how knowledge is gained and predictions are made for sustainability sciences and the global environment. AI methods, especially deep networks, have strong predictive skills yet are limited in interpretability and cannot alone answer specific scientific questions. Here we argue that the genes of AI, meaning its most transformative core ideas bringing forth favorable traits, can be absorbed into traditional modeling domains to give us new ways of making inquiries. For example, a genre of physics-informed machine learning, called “differentiable” modeling (DM, https://t.co/qyuAzYPA6Y), trains neural networks (NNs) with process-based equations (priors) together in one stage (called “end-to-end”) to benefit from the best aspects of both paradigms. DM can produce state-of-the-art predictions, inherently enable physical interpretation, extrapolate well in space and time, and leverage efficient AI computing infrastructure. We demonstrate the power of DM in hydrologic, river flow & transport, ecosystem, and water quality modeling, and use it to learn robust scientific answers from big data. As another example of AI “genes”, generative AI can capture the conditional joint distribution of multiple variables, depositing knowledge from all forms of observations and informing on scarcely observed processes – thus it can be leveraged to capture the coevolution of environmental variables. Tools with AI genes can serve as the basic infrastructure that democratizes access to reliable predictions and gives wings to solutions for some of the most elusive and important environmental modeling problems.

Bio: Chaopeng Shen is Professor in Civil Engineering at The Pennsylvania State University. He received the Ph.D. degree in environmental engineering from Michigan State University, East Lansing, MI, USA, in 2009. His PhD research focused on computational hydrology and he developed the hydrologic model Process-based Adaptive Watershed Simulator(PAWS), which was later coupled to the community land model to study the interactions between hydrology and ecosystem. He was a Post-Doctoral Research Associate with the Lawrence Berkeley National Laboratory, Berkeley, CA, USA, from 2011 to 2012, working on high-performance computational geophysics. His recent efforts focused on harnessing the big data and machine learning (ML) and physics-informed ML opportunities in advancing hydrologic predictions and understanding. As an early advocate for ML in geosciences, he has written technical, editorial, review and collective opinion papers on hydrologic deep learning to call to attention the emerging opportunities for scientific advances. He currently promotes differentiable modeling which seamlessly integrates neural networks and physics for knowledge discovery. In addition, his research interests also include floodplain systems, scaling issues, process-based hydrologic modeling, and hydrologic data mining. He is currently Editor of Journal of Geophysical Research – Machine Learning & Computation, an Associate Editor of the Water Resources Research and Chief Special Editor for Frontiers in AI: Water and AI.

The University of Washington is committed to providing access, equal opportunity and reasonable accommodation in its services, programs, activities, education and employment for individuals with disabilities. To request disability accommodation contact the Disability Services Office at least ten days in advance at: 206.543.6450/V, 206.543.6452/TTY, 206.685.7264 (FAX), or e-mail at dso@uw.edu