Machine Learning in Geospace

_

_

GEM 2022

Workshop Overview

What

We are back for a third year organizing the Machine Learning in Geospace session at the GEM 2022 Workshop. This year, we have three sessions. One will be filled with invited and submitted talks, one will focus on tools and applications, and another will focus on connections to industry.

Registration

Register for the in-person workshop at GEM website. Register for the virtual component through the VGEM website

When

The workshop begins on June 20 and ends on July 24.

Table 1: Workhop schedule. For more details see the Workshop webpage and our agenda below.

Event	Date	Day	Time (Hawaii)
Student Day	June 19	Sunday	8:00 AM - 3:30 PM
GEM Workshop	June 20 - 24	Monday - Friday	8:00 AM - 5:00 PM + Posters & Events
ML Session I	June 23	Thursday	10:30 AM - 12:00 PM
ML Session II	June 23	Thursday	1:30 PM - 3:00 PM
ML Session III	June 24	Friday	1:30 PM - 3:00 PM

Who

Table 2: Session organizers. *Student Organizers

Name		Email
Matthew Argall		matthew.argall@unh.edu
Jacob Bortnik		jbortnik@gmail.com
Wendy Carande		Wendy.Carande@lasp.colorado.edu
Doğa Ozturk		dsozturk@alaska.edu
Josh Rigler		erigler@usgs.gov
Jason Shuster		jason.r.shuster@nasa.gov
Brian Swiger*		swigerbr@umich.edu
Alexandra Wold*		Alexandra.Wold@colorado.edu

Agenda

Session 1: Submitted Talks

Submitted talks focus on applications of machine learning to research in geospace by the community. Talks will focus on the machine learning methodology in an educational sense.

Table 3: Speakers and presentation titles.

Duration	Name	Title
6+3 min	Boris Kramer	Learning low-dimensional dynamical systems operators for advection-dominated solar wind models
6+3 min	Xiangning Chu	Relativistic electron model in the outer radiation belt using a neural network approach
6+3 min	Heidi Nykyri	Applications of Information Theory to Unravel the origin of the Energetic Electron Microinjections
6+3 min	Matthew Blandin	Feature Importance: How a computer assesses our variables
6+3 min	Anthony Sciola (Slava Merkin)	Ring Current Plasma Pressure Reconstructed from Data-mined Magnetometer Measurements Embedded Within a Global MHD Model
6+3 min	Connor O'Brien	Neural Network Models of the Near-Earth Solar Wind and Magnetosheath
6+3 min	Andy Edmond	Unsupervised Clustering of Magnetospheric Dayside Data
6+3 min	Xin Cao	The Response of Equivalent Ionospheric Currents to the External Drivers via Machine Learning
6+3 min	Donglai Ma	Modeling the dynamic variability of sub-relativistic outer radiation belt electron fluxes using machine learning.
6+3 min	Jinxing Li	Modeling Ring Current Proton Fluxes Using Machine Learning and Van Allen Probe Measurements
		Open Discussion

Session 2: Tutorials and Talks

Tutorials will walk through topics on machine learning methods as applied to research. When possible, Jupyter Notebooks based on the tutorial will be made available to attendees for adaptation to their own research goals.

Table 3: Speakers and presentation titles.

Duration	Name	Title
15 min	Wendy Carande	Tutorial: Data science competitions and machine learning pitfalls
25 min	Ben Winjum	Tutorial: Docker & Datashader
15 min	Vishal Upendran	Tutorial: Spherical Harmonics
6+3 min	Adam Kellerman	A machine-learning model of electron phase-space density in Earth's radiation belts
6+3 min	Izzak Boucher	Global model of the electric potential using regularized linear regression and neural networks
6+3 min	Andong Hu	Multi-Hour-Ahead Magnetic Field Perturbations Prediction Using UQ-boost ML Methods and SuperMag data
6+3 min	Hadeem Farooki (Hyomin Kim)	A machine learning approach to understanding the physical properties associated with magnetic flux ropes in the solar wind at 1 AU
		Open Discussion

Session 3: Industry Panel & Social Mixer

The Industry Panel discussions brings together several people who have made the jump from academia to industry. Discussion topics include tailoring resumes, the interview process, turning your ideas into a business, and general life in industry vs. academia.

Table 3: Speakers and presentation titles.

Duration	Panelist	Affiliation
45 min	Kyle Bahr	Emergent Analytics
	Matt Gilson	Google, Argo
	Megan Cartwright	Netflix
45 min		Community Mixer

Bios

Kyle Bahr

Kyle Bahr is the owner and principal of Emergent Analytics, a management consulting company that specializes in bringing machine learning and other advanced analytics techniques to sustainable development problems. After completing his PhD. in Mining and Earth Systems Engineering at the Colorado School of Mines, he accepted a post-doctoral fellowship at the Missouri University of Science and Technology. Before founding Emergent Analytics, Kyle was an Assistant Professor for four years at Tohoku University in Sendai, Japan, where his research focused on resource development, especially in the area of the social license for geothermal energy. Kyle still conducts academic research and teaching through his associations with the University of Eastern Finland and the Colorado School of Mines. Kyle has worked on consulting projects on 6 continents. He has expertise in computer modelling, data science, and the application of machine learning methods to resource development challenges.

VGEM 2021

Workshop Overview

What

We are back for a second year organizing the Machine Learning in Geospace session at the VGEM 2021 Workshop. This year, we have two sessions. One will be filled with invited and submitted talks and the other will focus on organizing a Kaggle Competition within the community to address a large, inter-disciplinary science problem using machine learning and data science techniques. See our Schedule Overview here and our detailed schedule below.

Registration

Register for the Workshop by filling out this form. There is no fee for students. Others can pay the $25 registration fee here.

When

The workshop begins on July 25 and ends on July 30.

Table 1: Workhop schedule. For more details see the Workshop webpage and our agenda below.

Event	Date	Day	Time (Eastern)
Student Day	July 25	Sunday	1:00 - 5:00 PM
VGEM Workshop	July 26 - 30	Monday - Friday	11:00 AM - 4:30 PM + Posters
ML Session I	July 28	Wednesday	3:00 - 4:00 PM
ML Session II	July 30	Friday	1:00 - 2:30 PM

Who

Table 2: Session organizers.

Name		Email
Matthew Argall		matthew.argall@unh.edu
Jacob Bortnik		jbortnik@gmail.com
Josh Rigler		erigler@usgs.gov
Jason Shuster		jason.r.shuster@nasa.gov
Doğa Ozturk		dsozturk@alaska.edu
Wendy Carande		Wendy.Carande@lasp.colorado.edu

Agenda

Session I: Submitted Talks

Table 3: Speakers and presentation titles.

Duration	Name	Title
10 min	Matthew Argall	Session overview, decadal planning
10 min	Breakout Rooms	Ice Breaker
10 min	Ayris Narock	Ethical AI
6 min	Matthew Blandin	Prediction of Geomagnetic Field Disturbances across Alaska using Machine Learned LSTM Neural Networks
6 min	Sheng Huang	Recurrent neural network implementation of modelling global plasmaspheric density
6 min	Michael Coughlan	Using OMNI and SuperMag data to determine the risk of db/dt threshold exceedance with Long-Short Term Memory and Convolutional Neural Networks.
6 min	Chris Bard	Reconstructing MHD with Neural Networks
6 min	Anthony Saikin	NARX Neural Network Derivations of the Outer Boundary Radiation Belt Electron Flux
6 min	Eric Donovan	Applying ML to the THEMIS-ASI data set
30 min	Open Discussion

Session II: Kaggle Competition

Table 3: Speakers and presentation titles.

Duration	Name	Title
5 min	Matthew Argall	Session overview
15 min	Addison Howard	How to Design a Successful Kaggle Competition
5 min	Manoj Nair	NOAA's DataDriven MagNet Chellenge
5 min	Barbara Thompson	AI-Ready Datasets
5 min	Raphael Attie	Data classification tools
5 min	Wendy Carande	Getting Started with ML: Common Problems
20 min	Breakout Rooms	Designing the competition. * What problems can be addressed with an ML Competition? * What missions or datasets would be useful? * What resources are required to prepare for a successful competition? * What evaluation criteria will lead to a fair and meaningful winner? * Get the discussion started early by contributing to our Jam Board
15 min	Andrés Muñoz-Jaramillo Enrico Camporeale Shasha Zou	"What & How" A recap of the breakout sessions
20 min	Speakers & Panelists	Open Discussion on Kaggle Competition

Heliophysics Decadal Survey White Papers

Members of the Heliophysics Community are coordinating and organizing white papers on Machine Learning and Data Science for the Heliophysics Decadal Survey in order to increase their visibility and impact. Anyone interested in leading, contributing to, or supporting a white paper, please reach out to Matthew Argall matthew.argall@unh.edu for more details.

VGEM 2020

Workshop Overview

Student Day	Monday, July 20th, 2020 is Student Day
GEM Workshop	Tuesday - Thursday, July 21st - 23rd

Participants will need to register for free here, which is accessible through the general meeting website. Our session agenda is posted below.

Machine Learning Session

When

Thursday, July 23, 3:00 - 4:30pm Eastern

Session Conveners

Name		Email
Matthew Argall		matthew.argall@unh.edu
Jacob Bortnik		jbortnik@gmail.com
Josh Rigler		erigler@usgs.gov
Jason Shuster		jason.r.shuster@nasa.gov

Preface

We have worked with the GEM Student Representatives to incorporate a machine learning tutorial into their Student Day. Abby Azari will give a "getting started with machine learning" tutorial.

Session Overview and Theme

Beyond the mechanics of getting started, to fully take advantage of the power of machine learning, one needs to know what is possible. This is our session's theme -- to provide an overview of what is possible. This theme encompasses five focus areas:

1. Recent developments in machine learning that enable innovative scientific research
2. Aspects of machine learning or data science that are under-utilized in the field
3. Systems/chains of machine learning models and/or the integration of machine learning into numerical simulations
4. Possibilities and opportunities that exist outside of a purely academic career path enabled by machine learning
5. Bridging the SHINE, GEM, and CEDAR communities to facilitate multi-discipline collaborations

The hope is that recent developments in ML will inspire new avenues of research and career opportunities.

Presentations and Panel Discussion

Speakers and panelists are encouraged to address any of the focus areas 1-4 and to include elements that enable focus area 5.

ML Exploratory Committee

This session is chaired by members of a SHINE/GEM/CEDAR Machine Learning Exploratory Committee. The committee is composed of members of the SHINE, GEM, and CEDAR communities. Its goal is to pursue topics in Heliophysics Machine Learning that can best be addressed through a cross-disciplinary initiative. Anyone interested can contact us for more information.

Agenda

Session

Table 1: Schedule for the Machine Learning session at VGEM. The audience can direct additional questions to the speakers during the panel discussion.

Duration	Topic	Speaker
5 min	Opening remarks	Matthew Argall
15 min	Introduction to ML: "What is Possible?"	Jacob Bortnik (See Table 3)
7 min 3 min	Application(s): ML in GEM with relevance to SHINE/CEDAR	Enrico Camporeale
7 min 3 min	Application of Machine Learning to GNSS-Related Remote Sensing Tasks	Leo Liu
7 min 3 min	Predicting Solar Eruptions: What has ML done for us?	Sophie Murray
40 min	Open Panel Discussion	See Table 2

Panel Discussion

Table 2: After a brief self-introduction, panelists to field questions in an open discussion format.

Field	Panelist	Early Career	Theme
SHINE	Wendy Carande	Y	Pairing science problems with ML tools
SHINE	Barbara Thompson		NASA, Policy
GEM	Simon Wing		Information Theory
GEM	Xiangning Chu	Y	Global Reconstruction
CEDAR	Bharat Kunduri	Y	Insight Industry to academia
CEDAR	TBD	-	-
Planetary	Abby Azari	Y	Student tutorial Physics Informed ML Interpretable ML
Industry	Nick Bunch	Y	Academia to industry

Distributed Tutorial

Table 3: SHINE/GEM/CEDAR ML Exploratory Committee topics highlighting a subset of what is possible with machine learning. Each speaker provides an overview of the topic in ~1 minute and ~1 slide.

Speaker	Topic
Ryan McGranaghan	Community-wide inventory of assets and resources
Marcus Hughes	Exploration, visualization and handling of large observational and modeling data volumes
Hazel Bain	Understanding the potential enhancement of space weather prediction capabilities for operations and research
Barbara Thompson	Creation of AI-ready data sets and quality standards
Jacob Bortnik	Different approaches to implementing “Physics-informed learning” and “gray box models”
Josh Rigler	Compression, acceleration and emulation of models and data processing
Matthew Argall	Providing paths to publication, validation, and open-code practices