Citizen Scientists, NASA looking for Exoplanet volunteers

Science
#1

Help NASA find exoplanets, worlds beyond our solar system, through a newly launched website called Planet Patrol. This citizen science platform allows members of the public to collaborate with professional astronomers as they sort through a stockpile of star-studded images collected by NASA’s Transiting Exoplanet Survey Satellite (TESS).

“Automated methods of processing TESS data sometimes fail to catch imposters that look like exoplanets,” said project leader Veselin Kostov, a research scientist at NASA’s Goddard Space Flight Center in Greenbelt, Maryland, and the SETI Institute in Mountain View, California. “The human eye is extremely good at spotting such imposters, and we need citizen scientists to help us distinguish between the look-alikes and genuine planets.”

5 Likes
#2

Zooniverse is a great site. I’ve participated in a few of their projects. My favorite was transcribing handwritten notes from Civil War officers. Many concerned troop movements and strategies.

3 Likes
#3

That is really cool Tom. Miss you brother. There was a video on youtube recently of an old home in Virginia that was renovated and they found 100’s letters hidden under some floor boards, a few of them written by Robert E Lee.

I think for the project that I posted they want citizen scientist to label the data and eventually will feed it into Machine Learning Algos to help find exoplanets faster. I wouldn’t be surprised if in 15-20 years we have discovered 15,000 exoplanets and at least a dozen highly favorable candidates in the goldilocks zone, at the triple point of water.

2 Likes
#4

Great video of the letters. Wouldn’t it be so cool to find something like that!

1 Like
#5

This seems like a waste of time and resources. I can’t for the life of me figure out why it’s of use to anyone alive today to know that there’s a particular habitable planet 2 million light-years away.

We’ll have AI tools and quantum computers that can give us this information in a few fractions of a second anyway in a few dozen decades.

Bah! Humbug!

1 Like
#6

Not unless we can get humans to label data as a stepping stone. AI using ML must be taught what “we” consider important. The problem is so absurdly complex that we don’t have all the tools yet to connect all the dots. Somehow the human brain does this but we’ve only scratched the surface of how that works. You are right that we will have AI that can do this in fractions of a second but only because projects like this one paved the way by bootstrapping with pure human conignative ability. We need more projects like this, not less, if we want to get to your vision for AI.

The ironic part is that you can’t ask most ML models, “why did you think this was a cat?” Just like you can’t ask humans that question and get an accurate answer from multiple people. So we keep using our human learning model (brain) to bootstrap the machine learning models.

4 Likes
#7

I think they are preparing for the James Webb launch, there will probably be enormous amounts of data to sift through once it is operational. I wouldn’t be surprised if there are 100s of known earth like planets by 2030.

This is an extremely interesting video on imaging exoplanets in the future, and the great difficulty it presents us. If you are so inclined on the topic.

3 Likes