AI4ESP Workshop Oct-Dec 2021
The U.S. Department of Energy Artificial Intelligence for Earth System Predictability (AI4ESP) Workshop is currently taking place virtually with 17 sessions spanning 10/25/21 to 12/08/21.
Agenda:
Registration for the public portion:
Figure 1. Points of Contact for AI4ESP.
The U.S. Department of Energy has prioritized a strategy to explore novel new approaches to the science of Earth system predictability that can exploit developments in artificial intelligence, edge computing, 5G, exascale class computational architectures, and other emerging technologies. As part of this strategy, DOE launched the Artificial Intelligence for Earth System Predictability (AI4ESP) project in the fall of 2020, that now involves eight national laboratories plus academia as well as the private sector. Sponsored by the DOE’s Office of Biological and Environmental Research (BER) and Office of Advanced Scientific Computing Research (ASCR) within DOE’s Office of Science, this effort is a collaboration between DOE management led by Gary Geernaert and program managers from BER’s Earth and Environmental Systems Science Division (EESSD) and ASCR and Core Team members and Point of Contacts from DOE laboratories led by Nicki Hickmon (Figure 1). The goal is to promote a paradigm shift that combines AI with the Model-Experiment (MODEX) approach to research, in part by determining the most impactful applications along the observation-modeling-analysis-prediction continuum.
In response to an open solicitation of white papers from the research community, 156 White Papers were received by the AI4ESP Team that focused on development and application of AI methods in areas relevant to EESSD research. The clustering of white papers along common themes formed the basis of the ongoing AI4ESP Workshop. This workshop extends over a six-week period starting on October 25, 2021. Each week focuses on different topics of Earth system predictability research plus other cross-cutting themes, where each topic involves co-design with new and/or emerging AI research concepts. The major outcome of the workshop will be summarized in a high-level yet forward-looking report that will then be used by Office of Science to inform future priorities.
Session leads:
- Atmospheric Modeling – Chair: Ruby Leung ruby.leung@pnnl.gov
- Land Modeling – Chair: Beth Drewniak bbye@anl.gov
- Hydrology – Chair: Charuleka Varadharajan cvaradharajan@lbl.gov
- Watershed Science – Chair: Mavrik Zavarin zavarin1@llnl.gov
- Ecohydrology – Chair: Forrest Hoffman hoffmanfm@ornl.gov
- Aerosols & Clouds – Chair: Po-Lun Ma Po-Lun.Ma@pnnl.gov
- Climate Variability & Extremes – Chair: Maria Molina molina@ucar.edu
- Human Systems & Dynamics – Chair: Christa Brelsford brelsfordcm@ornl.go
- Coastal Dynamics, Oceans, & Ice – Chair: Matthew Hoffman mhoffman@lanl.gov
- Data Acquisition – Chair: Giri Prakash palanisamyg@ornl.gov
- Knowledge Discovery & Statistical Learning – Chair: Xingyuan Chen Xingyuan.Chen@pnnl.gov
- Neural Networks – Chair: Nathan Hodas nathan.hodas@pnnl.gov
- Surrogate Models & Emulators – Chair: Nathan Urban nurban@bnl.gov
- Knowledge Informed Machine Learning – Chair: Frank Alexander falexander@bnl.gov
- Hybrid Modeling – Chair: Jiali Wang jialiwang@anl.gov
- Explainable/Interpretable/Trustworthy AI – Chair: Line Pouchard pouchard@bnl.gov
- AI Architecture & CoDesign – Chair: Jim Wang ang@pnnl.gov
Organizers:
- The AI4ESP workshop organizers and Core Group: Nicki Hickmon, Forrest Hoffman, Haruko Wainwright, and Scott Collis.
