I thought the part at the end, where he talked about how astronomy hasn’t found its “grand challenges” yet, was really interesting. I wonder what the astro equivalent of the protein folding question would be like…
I caught 10min of the premier, but seems like a super interesting guy with a great perspective from the trenches of what its like doing ML for astro! Need to go back and re-watch
I think the episode is pretty information-dense, and I definitely re-listened to a few parts in order for things to sink in. There’s a transcript that you might find useful!
About 12, 13 years ago, we were actually dealing with lots of images coming off of telescopes from the ground. The normal behavior when you get lots of data had been — and in many circles still is — just hire more grad students to look at the data.
It’s baffling how far Computer Vision has come and the impact it’s having on the world.
“just hire more grad students to look at the data” — Grad Student Descent discussed on the Gradient Dissent Podcast, is this a first!?