TrustWorthy AI: Advances, Applications, and Assessing Uncertainties
TrustWorthy AI: Advances, Applications, and Assessing Uncertainties
Session ID#: 123478
Session Description:
With the advances of machine learning and artificial intelligence in the past couple decades, the applications of these models have substantially increased. Unfortunately, many modelers blindly apply these methods in areas that they have no expertise, or infer causality based on insufficient amount of data. The complexity of machine learning methods (especially deep learning), also makes inference of bases of the decision making of these models very difficult.
We welcome studies that investigate the uncertainties of machine learning and artificial intelligence in earth sciences and create new methods for evaluating the reliability of machine learning models.
Index Terms:
0299 General or miscellaneous [GEOHEALTH]
0550 Model verification and validation [COMPUTATIONAL GEOPHYSICS]
0555 Neural networks, fuzzy logic, machine learning [COMPUTATIONAL GEOPHYSICS]
4499 General or miscellaneous [NONLINEAR GEOPHYSICS]
Primary Convener: Mohammadhassan Mohegh, George Washington University, Public Health, Washington, DC, United States
Conveners: Mohammadhassan Mohegh1, George A Ban-Weiss1 and Mo Chen2, (1)University of Southern California, Civil and Environmental Engineering, Los Angeles, CA, United States(2)University of Southern California, Los Angeles, CA, United States
Primary Liaison: Mohammadhassan Mohegh, University of Southern California, Civil and Environmental Engineering, Los Angeles, CA, United States
See more of: GeoHealth