Embeddings in Context: Tools for Geospatial Problem Solving

https://is1-ssl.mzstatic.com/image/thumb/Podcasts221/v4/60/d2/2c/60d22c39-1991-c54a-c08e-898d158a128a/mza_241129215262153427.jpg/600x600bb.jpg

Earthsight

cr458

4 episodes

6 days ago

Chris and Krishna share big thoughts about the geospatial industry. Chris is a data scientist currently working in weather, but has previously worked at Los Alamos National Lab, The Earth Genome, Gro Intelligence and Demeter Labs. At those places he used earth observation data to solve problems such as crop mapping, yield prediction, change detection etc... Krishna is a data journalist who uses satellite imagery for news coverage. Previously he worked at Descartes Labs, Impact Observatory, The Earth Genome, Ceres Imaging and more. Krishna is an adjunct professor at The Cooper Union.

Science

RSS

All content for Earthsight is the property of cr458 and is served directly from their servers with no modification, redirects, or rehosting. The podcast is not affiliated with or endorsed by Podjoint in any way.

Science

https://d3t3ozftmdmh3i.cloudfront.net/staging/podcast_uploaded_nologo/44375827/44375827-1757368220433-a8cd90d7df858.jpg

Embeddings in Context: Tools for Geospatial Problem Solving

Earthsight

59 minutes 44 seconds

2 months ago

Embeddings in Context: Tools for Geospatial Problem Solving

Chris and Krishna discuss their experiences working with geospatial embeddings, search and remote sensing in general, and how they think through problems.

Timestamps sustainably sourced and hand crafted.

(0:00) Intro

(2:30) Background, earth observation and existing algorithms and the embedding fallacy.

(9:49) Validation, how to solve problems, seductive embeddings, the museum of cool demos.

(19:22) What are embeddings?

(21:00) Search vs embeddings, the development of Earth Index.

(25:00) Expert embeddings: the dumbest embeddings that work.

(30:25) Computer vision embeddings/ImageNet. Experiments in pre-training, how to spend $40,000 in cloud credits.

(37:22) Should we pre-train on satellite imagery? DINO v3

(44:30) What properties should embeddings have for search? Recall vs Precision/Landcover+

(50:00) Real world consequences of mapping

(54:00) Future outlook, Krishna is jaded