IN11E:
Recent Advances in Deep Learning and Data Analytics in Earth, Atmospheric, and Planetary Sciences II: Deep Learning to Solve Geoscience Challenges

Submit an Abstract to this Session



Session ID#: 33668

Session Description:
Machine learning methods have revolutionized many aspects of our daily life and are poised to revolutionize far more. Some areas of Earth and space science have long established histories of use, but most have seen only limited application. As Deep Learning and Deep Belief emerge as potentially benefiting areas such as hydrometerological prediction, they also gain more visibility in other fields (and even TV programs). The incorporation of uncertainty, physics and domain knowledge, and space-time dependence in highly scalable Deep Learning architectures may provide a base to model and learn abstract representations of complex earth science datasets. This session seeks papers in theory, adoptions, practice, anecdotal experiences, use cases, and lessons learned in applying machine learning, deep learning and deep belief in Earth and space science.
Primary Convener:  Jens F Klump, CSIRO Earth Science Resource Engineering, Perth, Australia
Conveners:  Michael M Little, NASA Headquarters, Earth Science Technology Office, Washington, DC, United States, Youzuo Lin, Los Alamos National Laboratory, Los Alamos, United States and Udit Bhatia, Northeastern University, Boston, MA, United States
Chairs:  Youzuo Lin, Los Alamos National Laboratory, Los Alamos, United States and Jens F Klump, CSIRO Mineral Resources, Perth, Australia
OSPA Liaison:  Michael M Little, NASA Headquarters, Washington, DC, United States

Cross-Listed:
  • A - Atmospheric Sciences
  • EP - Earth and Planetary Surface Processes
  • H - Hydrology
  • S - Seismology
Index Terms:

1914 Data mining [INFORMATICS]
1916 Data and information discovery [INFORMATICS]
1942 Machine learning [INFORMATICS]
1968 Scientific reasoning/inference [INFORMATICS]

Abstracts Submitted to this Session:

Machine Learning of Fault Friction (257361)
Paul A Johnson1, Bertrand Rouet-Leduc2, Claudia Hulbert3, Chris Marone4 and Robert A Guyer2, (1)Los Alamos National Laboratory, Los Alamos, NM, United States, (2)Los Alamos National Laboratory, Earth and Environmental Sciences, Los Alamos, NM, United States, (3)Geolabe, Los Alamos, United States, (4)Penn State University, Department of Geosciences, University Park, PA, United States
Large-scale Labeled Datasets to Fuel Earth Science Deep Learning Applications (229203)
Manil Maskey1, Rahul Ramachandran1 and Jeffrey Miller2, (1)NASA Marshall Space Flight Center, Huntsville, United States, (2)Zelo Buon Persico, Italy
Using Deep Learning for Tropical Cyclone Intensity Estimation (240489)
Jeffrey Miller, University of Alabama in Huntsville, Huntsville, AL, United States, Manil Maskey, NASA Marshall Space Flight Center, Huntsville, United States and Todd Berendes, The University of Alabama in Huntsville, Information Technology and Systems Center, Huntsville, AL, United States
Simplify and Accelerate Earth Science Data Preparation to Systemize Machine Learning (Invited) (214755)
Kwo-Sen Kuo1,2, Michael Lee Rilee3 and Amidu Oloso3, (1)NASA Goddard Space Flight Center, Greenbelt, United States, (2)Bayesics, LLC, Bowie, MD, United States, (3)NASA Goddard Space Flight Center, Greenbelt, MD, United States
New Techniques for Deep Learning with Geospatial Data using TensorFlow, Earth Engine, and Google Cloud Platform (266920)
Matthew Hancher, Google, Mountain View, CA, United States and The Google Earth Engine Team

See more of: Earth and Space Science Informatics