EESS talk on "Atmospheric Physics-Guided Machine Learning: Towards Physically-Consistent, Data-Driven, and Interpretable Models of Convection"
Event details
| Date | 20.09.2022 |
| Hour | 12:15 › 13:15 |
| Speaker | Dr Tom Beucler, Assistant professor, Institute of Earth Surface Dynamics, UNIL |
| Location | Online |
| Category | Conferences - Seminars |
| Event Language | English |
Abstract:
Data-driven algorithms, in particular neural networks, can emulate the effect of unresolved processes in coarse-resolution climate models if trained on high-resolution simulation data. However, they may violate key physical constraints and make large errors when evaluated outside of their training set. I will share progress towards overcoming these two challenges in the case of machine learning the effect of subgrid-scale convection and clouds on the large-scale climate. First, physical constraints can be enforced in neural networks, either approximately by adapting the loss function or to within machine precision by adapting the architecture. Second, as these physical constraints are insufficient to guarantee generalizability, I additionally propose to physically rescale the inputs and outputs of machine learning algorithms to help them generalize to unseen climates. Overall, these results suggest that explicitly incorporating physical knowledge into data-driven models of climate processes may improve their consistency, stability, and ability to generalize across climate regimes.
Short biography:
Tom Beucler is an assistant professor of environmental data science at the University of Lausanne in Switzerland. He recently started a lab specifically dedicated to the intersection of atmospheric physics and machine learning, with the goal of improving our understanding of atmospheric dynamics and assisting weather and climate predictions. For that purpose, his research group combines physical theory, computational science, statistics, numerical simulations, and observational analyses. Before that, Tom studied the interaction of tropical storms, radiation, and atmospheric water as part of his PhD at MIT. As a postdoc and project scientist at Columbia and UC Irvine, he investigated how to best integrate physical knowledge into neural-network representations of convection for climate modeling, which will be the theme of today’s webinar.
Data-driven algorithms, in particular neural networks, can emulate the effect of unresolved processes in coarse-resolution climate models if trained on high-resolution simulation data. However, they may violate key physical constraints and make large errors when evaluated outside of their training set. I will share progress towards overcoming these two challenges in the case of machine learning the effect of subgrid-scale convection and clouds on the large-scale climate. First, physical constraints can be enforced in neural networks, either approximately by adapting the loss function or to within machine precision by adapting the architecture. Second, as these physical constraints are insufficient to guarantee generalizability, I additionally propose to physically rescale the inputs and outputs of machine learning algorithms to help them generalize to unseen climates. Overall, these results suggest that explicitly incorporating physical knowledge into data-driven models of climate processes may improve their consistency, stability, and ability to generalize across climate regimes.
Short biography:
Tom Beucler is an assistant professor of environmental data science at the University of Lausanne in Switzerland. He recently started a lab specifically dedicated to the intersection of atmospheric physics and machine learning, with the goal of improving our understanding of atmospheric dynamics and assisting weather and climate predictions. For that purpose, his research group combines physical theory, computational science, statistics, numerical simulations, and observational analyses. Before that, Tom studied the interaction of tropical storms, radiation, and atmospheric water as part of his PhD at MIT. As a postdoc and project scientist at Columbia and UC Irvine, he investigated how to best integrate physical knowledge into neural-network representations of convection for climate modeling, which will be the theme of today’s webinar.
Practical information
- General public
- Free
- This event is internal
Organizer
- EESS - IIE
Contact
- Prof. D.Andrew Barry, ECOL