Session 2B Applying Uncertainty Quantification Methods to Environmental Artificial Intelligence Models

Monday, 29 January 2024: 10:45 AM-12:00 PM
338 (The Baltimore Convention Center)
Host: 23rd Conference on Artificial Intelligence for Environmental Science
Cochairs:
Eric D. Loken, OU/CIMMS, Norman, OK and Ryan Lagerquist

We invite abstracts that have experimented with AI uncertainty quantification methods applied to a variety of environmental science topics. In particular, we encourage abstracts highlighting what comes next after the uncertainty quantification and how to use the information gathered in model application.

Papers:
10:45 AM
2B.1
Machine-Learned Uncertainty Quantification Is Not Magic: Lessons Learned from Emulating Radiative Transfer with ML
Ryan A. Lagerquist, CIRA and NOAA/ESRL/GSL, Boulder, CO; and D. D. Turner, J. Q. Stewart, and I. Ebert-Uphoff

11:00 AM
2B.2
Evidential Deep Learning: Enhancing Predictive Uncertainty Estimation for Earth System Science Applications
Gabrielle Gantos, NCAR, Boulder, CO; and J. Schreck, D. J. Gagne II, C. Becker, W. E. Chapman, D. Kimpara, E. Kim, T. Martin, M. J. Molina, J. T. Radford, B. Saavedra, J. Willson, and C. D. Wirz

11:15 AM
2B.3
Explaining the Sources of Uncertainty in Machine Learning Winter Precipitation-Type Predictions
Charlie Becker, NCAR, Boulder, CO; and D. J. Gagne II, Ph.D., J. Schreck, G. Gantos, T. Martin, D. Kimpara, B. Saavedra, J. Willson, E. Kim, J. Demuth, C. D. Wirz, N. P. Bassill, K. J. Sulia, and A. McGovern

11:30 AM
2B.4
Uncertainty Estimation of Wind Gust Predictions Using the Deep Evidential Model
Israt Jahan, University of Connecticut, Storrs, CT; and M. Astitha, J. Schreck, and D. J. Gagne II

11:45 AM
2B.5
Uncertainty Quantifications of the Onset and Offset of Cold-Stunning Events Using AI Ensemble Methods
Hector Miguel Marrero Colominas, Texas A&M University-Corpus Christi, Corpus Christi, TX; NSF Artificial Intelligence Institute (AI2ES), Corpus Christi, TX; and M. Vicens-Miquel, P. E. Tissot, J. Woodall, C. Duff, and B. Colburn

- Indicates paper has been withdrawn from meeting
- Indicates an Award Winner