A recently developed AI model by NASA has helped identify nearly 7,000 exoplanets in data collected by the agency's Transiting Exoplanet Survey Satellite (TESS) and Kepler missions. The new model, called ExoMiner++, uses artificial intelligence to analyze TESS data and flag potential exoplanet candidates.
This is not a significant departure from the work done earlier by NASA, where an AI-powered software package named ExoMiner was used in 2021 to validate over 370 new exoplanets from Kepler data. Since then, the team has created an updated version of the model called ExoMiner++, which can now analyze both Kepler and TESS data.
The new algorithm worked on the vast dataset collected by TESS, and identified 7,000 potential exoplanet candidates in just one run. This is a testament to the effectiveness of AI in analyzing large datasets like this.
With its ability to sift through observations of possible transits to predict which ones are caused by exoplanets, ExoMiner++ promises to further accelerate scientific discovery. NASA's chief science data officer Kevin Murphy noted that open-source software such as ExoMiner accelerates scientific progress.
The availability of the model for free on GitHub allows researchers to access it and use it to hunt for planets in TESS’s growing public data archive. This is an essential step forward, given that open data and code are considered crucial pillars of gold-standard science.
As NASA gears up for its upcoming Nancy Grace Roman Space Telescope mission, which will capture tens of thousands of exoplanet transits, the advances made with ExoMiner models could potentially aid in the discovery of more planets from Roman data.
This is not a significant departure from the work done earlier by NASA, where an AI-powered software package named ExoMiner was used in 2021 to validate over 370 new exoplanets from Kepler data. Since then, the team has created an updated version of the model called ExoMiner++, which can now analyze both Kepler and TESS data.
The new algorithm worked on the vast dataset collected by TESS, and identified 7,000 potential exoplanet candidates in just one run. This is a testament to the effectiveness of AI in analyzing large datasets like this.
With its ability to sift through observations of possible transits to predict which ones are caused by exoplanets, ExoMiner++ promises to further accelerate scientific discovery. NASA's chief science data officer Kevin Murphy noted that open-source software such as ExoMiner accelerates scientific progress.
The availability of the model for free on GitHub allows researchers to access it and use it to hunt for planets in TESS’s growing public data archive. This is an essential step forward, given that open data and code are considered crucial pillars of gold-standard science.
As NASA gears up for its upcoming Nancy Grace Roman Space Telescope mission, which will capture tens of thousands of exoplanet transits, the advances made with ExoMiner models could potentially aid in the discovery of more planets from Roman data.
They're talking about 7k new exoplanets and I'm like 'wow' 
NASA's AI model is super smart! 
It's gonna make finding planets even easier for scientists. I think this is a big deal, not just for space exploration but also for helping us learn more about the universe 
it was so weird having my perspective all messed up like whoa 
anyway back to space stuff yeah the exoplanet thing is wild but what if we found life elsewhere and had to deal with intergalactic bureaucracy 
just think about it, "can you please fill out form 345-B to confirm planet Y's habitability?" 
can you believe we're on the cusp of discovering so many new planets already! I mean 7k is a crazy number and it's all thanks to this AI model that's making life easier for scientists. I think what's really cool about ExoMiner++ is that it's open source, which means anyone can access and use it. That's gonna accelerate the whole exoplanet hunting process so much. It's like, who knows how many new worlds we'll discover next? The fact that NASA is pushing for open data and code is huge too, because transparency and collaboration are key to making real progress in science. Now let's get ready to see what other secrets the universe has in store for us! 
yaaas i'm so hyped about this new AI model, 7000+ exoplanets is a huge deal! and that it's open-sourced on GitHub is a total game changer for researchers. can't wait to see what kind of discoveries they'll make with all that data 
I'm all about open-source software, it's like they say more data + code = progress 
can't wait to see what other discoveries are made possible by this tech 
. It's crazy to think that they were able to identify over 7,000 potential exoplanets in just one run - that's insane! I'm also hyped that it's open-source and available on GitHub 
. Overall, this is some cool stuff that could lead to some amazing discoveries! 
anyway back to space... 7k exoplanets sounds crazy but what's the deal with all these new discoveries? are we just gonna keep finding more planets or is there something else going on? 
also can someone pls make a vid of ExoMiner++ in action? that would be so cool