Participants who are planning to join in-person in Glasgow, please note the deadline to secure your accommodation through Glasgow City Marketing Bureau is 13 September 2021

t05 - Advances in machine learning and computer vision applications for space weather

Thomas Chen (Academy for Mathematics, Science, and Engineering)

TDM type: 4.Other

In this facilitated open discussion, attendees will be provided with the opportunity for high-level engagement. The conveners will call on attendees with hands raised (whether it1s virtual or in person) to contribute, one after another. There will be opportunities for debate and counterarguments within these discussions. The focus of the TOM is machine learning and computer vision applications for space weather. These range all the way from something as simple as linear regression to more complex concepts like random forests support vector machines, convolutional neural networks, and generative adversarial networks. The overall session will be broken down into four main topical sections to better structure the discourse. These sections at least preliminarily will be (1) ground-based observation, (2) satellite-based observation, (3) applications relating to effects on humans and (4) linking to terrestrial weather. Finally, there we be a "special session" at the end that focused specifically on computer vision applications, which are unique as they represent relatively recent developments in the last decade relating to deep learning and neural networks and their respective applications to imagery data especially. While computer vision will certainly be discussed in other sections as well, there will be a more directed focus on this topic at the end. After all discussions have ended, there will be free time where attendees can intermingle and potentially develop interdisciplinary collaborations that might even last beyond this one TOM.

Key questions and issues:

  • For which research questions are traditional statistical models preferable to more recent Al methods and vice versa, and why?
  • Are there ethical/legal constraints to consider when discussing the deployment of ML- or computer vision-based systems into the real world?
  • Is there enough 11big data 11 to tackle many of the research questions we wish to tackle with ML/CV?
  • Are Al techniques more useful in ground-based or satellite-based domains, and why?

Approaching deadlines:

Hotel booking deadline:

13 September 2021

Registration deadline:

01 October 2021

Click to see the list of key dates

The ESWW logo was designed by Wim Vander Putten and its continued use is courtesy of the STCE, Belgium.