Skip to content

GFDL Events & Seminars

Visitors without GFDL affiliation attending seminars or other organized events must present government or university issued photo ID or two other forms of identification to gain access to the facility. If an acceptable ID cannot be provided, the Visitor will not be allowed access. If access is granted, the Visitor must sign in and be given a Visitor Badge. The Visitor Badge expires immediately after the seminar.

June 24, 2020

calendar_today Hybrid modeling: best of both worlds ?

person Lunchtime Seminar - Virtual Lunchtime Seminar Series - Pierre Gentine (Columbia University- Dept of Earth and Environmental Engineering)

access_time 12:00 pm - 1:00 pm

place Location: Smagorinsky Seminar Room

n recent years, we have witnessed an explosion in the applications of machine learning, especially for environmental problems.Yet for broader use, those algorithms may need to respect exactly some physical constraints such as the conservation of mass and energy. In addition, environmental applications (e.g. drought, heat waves) are typically focusing on extremes and on out-of-sample generalization rather than on interpolation. This can be a problem for typical algorithms, which interpolate well but have difficulties extrapolating. I will here show how a hybridization of machine learning algorithms, imposing physical knowledge within them, can help with those different issues and offer a promising avenue for climate applications and process understanding. BIO Pierre Gentine is an associate professor in Earth and Environmental Engineering at Columbia. He studies the terrestrial water and carbon cycles and their changes with climate change. Pierre Gentine is recipient of the NSF, NASA and DOE early career awards, as well as the American Geophysical Union Global Environmental Changes Early Carrer and American Meteorological Society Meisinger award.