Bruno Sánchez-Andrade Nuño

On the agenda

1. Finetuning Earth Observation (EO) Foundation Models Workshop

   May. 16 '24, 03:00pm

   ---

   Workshop

   ---

   The North Hub

   Earth Observation Foundation Models are a potential game-changer for climate and sustainable development. Distilling and synthesizing vast amounts of earth observation (EO) data is a powerful way to provide useful inputs for numerous downstream applications such as forest carbon measurement, disaster detection, or crop monitoring at a lower cost.

   This hands-on workshop will introduce participants to applying Earth Observation (EO) foundation models in their own work. Participants will practice finetuning existing EO foundation models for specific sustainability applications, such as land cover, crop harvests, or flooding. The workshop will cover foundation models such as 1) Clay, 2) IBM and NASA's watsonx.ai geospatial foundation model, and 3) potentially a model from the European Space Agency’s FAST-EO project. With support from Amazon Web Services, participants will use SageMaker notebooks to do this exploration at no cost.

   The workshop will be run as part of SatSummit in Washington, D.C. on the 16th and 17th of May, 2024. If you’re already attending SatSummit, please let us know you’re interested in attending the workshop by completing this form. If you'd like to attend the workshop independently of the broader event, you can purchase a AI Finetuning Training Only ticket.

   Facilitated by: Clay

2. Large Earth Foundation Models

   May. 17 '24, 09:00am

   ---

   Session

   ---

   The Forum (South)

   The emergence of Generative AI is perhaps the greatest technology advancement since the last SatSummit. Large Earth Models are emerging that draw from similar approaches as Large Language Models (LLMs). These Large Earth Models can potentially revolutionize how we derive earth insights. This session will review the Foundation Models approach, which differs from prior AI approaches, and share several promising implementations.