Artificial Intelligence for Earth System Science (AI4ESS) Summer School 2020

Date and Time

The AI4ESS Summer School was held virtually June 22-26, 2020 by the National Center for Atmospheric Research (Boulder, Colorado, USA).

***Due to COVID-19 travel restrictions, the AI4ESS summer school was held virtually.***

Our schedule is in Mountain Daylight Time (Denver, Colorado, USA). Please use a time zone converter (like this one) if you are unsure what time it equals to in your own time zone.

Follow Along with the Hackathon

If you are unable to participate in the hackathon, you may follow along with the problems on your own with the Google colab links in GitHub. Links to the completed team notebooks are available in GitHub

Description

The AI4ESS Summer School is a week long interactive course covering essential and cutting edge topics in artificial intelligence, machine learning, and deep learning that are relevant for Earth System Science problems. Each morning, students will participate in lectures from leading researchers at the intersection of AI and ESS on AI fundamentals, ESS AI applications, and emerging methods. Afternoons will feature interactive breakout sessions where students will work in teams to solve challenge problems with real Earth System data and the AI techniques they have studied during the week. There will also be opportunities to network with the instructors and other students and present a poster about your research.

Goals

The goals of the summer school are as follows:

  1. Learn about the fundamentals of data processing, machine learning and deep learning algorithms, evaluation, and interpretation.
  2. See how machine learning systems have been developed for a range of Earth System Science applications.
  3. Develop hands-on experience with machine learning techniques covered in the course on real-world Earth System Science datasets.
  4. Investigate new and emerging methods for ESS machine learning.
  5. Network with fellow students and instructors.

Program

The schedule for the summer school can be downloaded here

Links to slides and recordings in menu panel to the left. 

Recommended Resources

Please see this PDF for recommended reading before the summer school.

Please see this PDF for other resources, including events and books recommended by our speakers. 

Confirmed Instructors

Amy McGovern, University of Oklahoma
Chaopeng Shen, Penn State
Claire Monteleoni, University of Colorado Boulder
David Hall, NVIDIA
David John Gagne, NCAR
Dorit Hammerling, Colorado School of Mines
Imme Ebert-Uphoff, Colorado State University/CIRA
Jebb Stewart, NOAA ESRL
Karthik Kashinath, Lawrence Berkeley National Laboratory
Katie Dagon, NCAR
Mike Pritchard, UC Irvine
Mustafa Mustafa, Lawrence Berkeley Lab
Pierre Gentine, Columbia University
Ryan Lagerquist, University of Oklahoma
Sue Ellen Haupt, NCAR

Certificate of Participation

If you would like a certificate of participation, please email taysia@ucar.edu. You must have registered for AI4ESS to receive a certificate, even if just for lectures. 

Cost

There will be no registration fee for the summer school. 

Summer School Program Coordinators

David John Gagne, NCAR
Karthik Kashinath, Lawrence Berkeley National Laboratory
Rich Loft, NCAR

Please direct questions about the program to David John Gagne (dgagne@ucar.edu)

Summer School Administrator

Taysia Peterson (taysia@ucar.edu)

Code of Conduct

UCAR and NCAR are committed to providing a safe, productive, and welcoming environment for all participants in any conference, workshop, field project or project hosted or managed by UCAR, no matter what role they play or their background. This includes respectful treatment of everyone regardless of gender, gender identity or expression, sexual orientation, disability, physical appearance, age, body size, race, religion, national origin, ethnicity, level of experience, political affiliation, veteran status, pregnancy, genetic information, as well as any other characteristic protected under state or federal law. 

All participants (and guests) are required to abide by this Code of Conduct. This Code of Conduct is adapted from the one adopted by AGU, complies with the new directive from the National Science Foundation (NSF) and applies to all UCAR-related events, including those sponsored by organizations other than UCAR but held in conjunction with UCAR events, in any location throughout the world. 

The full Code of Conduct document can be found here.

Sponsored by

National Science Foundation Logo

NCAR Logo

UCAR Logo

Berkeley Lab Logo

VAISALA Logo

Cloud Computing

AWS Logo

Presentation Slides

If you re-use any of the follow material, you are agreeing to acknowledge the authors of the presentations. 

 

Monday, June 22, 2020

Building a Strong Foundation: Defining ML Problems and Preprocessing - David John Gagne - NCAR

Machine Learning Fundamentals - Dorit Hammerling - Colorado School of Mines

Decision Trees and Ensembles - Ryan Lagerquist - University of Oklahoma

Hackathon Intro - David John Gagne - NCAR

 

Tuesday, June 23, 2020

Convolutional Neural Networks - Karthik Kashinath - Lawrence Berkeley National Laboratory

Recurrent Neural Networks and LSTMs - Chaopeng Shen - Penn State University

Deep Learning Architectures - David Hall - NVIDIA

 

Wednesday, June 24, 2020

ML in Weather Forecasting Systems - Sue Ellen Haupt - NCAR

ML for Segmentation of Atmospheric Phenomena - Jebb Stewart - NOAA ESRL

ML Emulators in Land Surface Models - Katie Dagon - NCAR

 

Thursday, June 25, 2020

Peering Inside the Black Box of Machine Learning for Earth Science - Part 1 - Amy McGovern - University of Oklahoma

Peering Inside the Black Box of Machine Learning for Earth Science - Part 2 - Imme Ebert-Uphoff - Colorado State University/CIRA

ML Parameterization - Mike Pritchard - UC Irvine

 

Friday, June 26, 2020

Generative Models - Mustafa Mustafa - Lawrence Berkeley National Laboratory

Physics-Guided ML - Pierre Gentine - Columbia University

Deep Unsupervised Learning for Climate Applications - Claire Monteleoni - University of Colorado Boulder

GOES - Team Slides

GECKO - Team Slides

ELNINO - Team Slides

HOLODEC - Team Slides

Presentation Recordings

When available, they will be posted individually below and added to this YouTube playlist.

 

Monday, June 22, 2020

AI4ESS Summer School Intro - Rich Loft, NCAR

Building a Strong Foundation: Defining ML Problems and Preprocessing - David John Gagne - NCAR

Machine Learning Fundamentals - Dorit Hammerling - Colorado School of Mines

Decision Trees and Ensembles - Ryan Lagerquist - University of Oklahoma

Panel Discussion

Hackathon Intro - David John Gagne - NCAR

 

Tuesday, June 23, 2020

Convolutional Neural Networks - Karthik Kashinath - Lawrence Berkeley National Laboratory

Recurrent Neural Networks and LSTMs - Chaopeng Shen - Penn State University

Deep Learning Architectures - David Hall - NVIDIA

Panel Discussion

Hackathon Update - David John Gagne - NCAR

Wednesday, June 24, 2020

ML in Weather Forecasting Systems - Sue Ellen Haupt - NCAR

ML for Segmentation of Atmospheric Phenomena - Jebb Stewart - NOAA ESRL

ML Emulators in Land Surface Models - Katie Dagon - NCAR

Panel Discussion

Hackathon Update - David John Gagne, NCAR

 

Thursday, June 25, 2020

Peering Inside the Black Box of Machine Learning for Earth Science - Part 1 - Amy McGovern - University of Oklahoma

Peering Inside the Black Box of Machine Learning for Earth Science - Part 2 - Imme Ebert-Uphoff - Colorado State University/CIRA

ML Parameterization - Mike Pritchard - UC Irvine

Panel Discussion

Hackathon Update - David John Gagne, NCAR

 

Friday, June 26, 2020

Generative Models - Mustafa Mustafa - Lawrence Berkeley National Laboratory

Physics-Guided ML - Pierre Gentine - Columbia University

Deep Unsupervised Learning for Climate Applications - Claire Monteleoni - University of Colorado Boulder

Panel Discussion

Hackathon Presentations - Ankur Mahesh (Lawrence Berkeley Lab), David John Gagne, Charlie Becker, Gabrielle Gantos (NCAR)

Summer School Application

Our summer school is in-session. You can watch lectures here: https://operations.ucar.edu/live-AI4ESS