When MIT launched the Stephen A. Schwarzman College of Computing this fall, one of the goals was to encourage more innovation in computing across all schools at MIT. Researchers are already expanding beyond the traditional applications of computer science and using these techniques to advance a range of scientific fields, from cancer treatment to anthropology to design – and to the discovery of new planets.
Computing has already been useful for Transit Exoplanet Survey Satellite (TESS), a NASA-funded mission led by MIT. Launched from Cape Canaveral in April 2018, TESS is a satellite that takes images of the skies as it moves around the Earth. These images can help researchers find planets erupting stars outside our sun, called exoplanets. This work, now halfway complete, will reveal more about the other planets within what NASA calls a “solar neighborhood.”
“TESS has just completed the first of its main two-year mission, exploring the southern night sky,” said Sara Seager, an astrophysicist and planetary expert at MIT and deputy director of science for TESS. “TESS has discovered over 1,000 planetary candidates and around 20 proven planets, some in multi-planet systems.”
While TESS has enabled intrusive detections to date, locating these exoplanets is not a simple task. TESS collects images of more than 200,000 distant stars, saving an image of these planets every two minutes, as well as saving an image of a giant moon of the sky every 30 minutes. Seager says every two weeks, which is how long it takes the satellite to orbit the Earth, TESS sends about 350 gigabytes of data (once unmatched) to the Earth. While Seager says this is not as much data as people would expect (the Macbook Pro 2019 has up to 512 gigabytes of storage), analyzing the data is a consideration on many complex factors.
Seager, who says she has a longstanding interest in how computing can be used as a tool for science, began talking about the project with Victor Pankratius, who was a leading research scientist at the Institute. Kavli for Astrophysics and Space Research at MIT, who is now director and head of global software engineering at Bosch Sensortec. A trained computer scientist, Pankratius says, after arriving at MIT in 2013, he began to think about scientific fields that produce big data, but have not yet fully benefited from computing techniques. After talking to astronauts like Seager, he learned more about the data their instruments collect and became interested in using computer-assisted detection methods to detect exoplanets.
“The universe is a big place,” says Pankratius. “So I think it’s a good idea to reduce what we have on the computer science side. ”
The basic idea behind the TESS mission is that, like our own solar system, in which the Earth and other planets revolve around a central star (the sun), there are other planets outside our solar system that revolving around different stars. The images TESS collects emit light loops – data that shows how the star ‘s brightness changes over time. Researchers are analyzing these light loops to find droplets in brightness, which may indicate that a planet is passing in front of the star and blocking some of the space. her light for a while.
“Every time a planet orbits, you would see this brightness go down,” says Pankratius. “It’s almost like a heartbeat.”
The problem is that not all declines in brightness are caused by a passing planet. Seager says that machine learning is currently entering the “triage” phase of their TESS data analysis, helping them to differentiate between potential planets and other objects. ‘may cause a decrease in brightness, such as variable stars, which naturally change in their brightness, or the sound of an instrument.
Analysis of planets that pass through triage is still being done by scientists who have learned how to read light loops. But the team is now using thousands of eye-catching light loops to teach cloud networks how to recognize exoplanet transmissions. Computing helps them reduce which light loops they should examine in more detail. Liang Yu Ph.D. ’19, a recent physics graduate, built on existing code to write the machine learning tool the team is now using.
While helpful for importing the most relevant data, Seager says machine learning can’t yet be used to just find exoplanets. “We still have a lot of work to do,” she says.
Pankratius agrees. “All we want to do is create computer-aided search systems that do this for everyone [stars] all the time, ”he says. “You just want to push a button and say, show me everything. But for now there are still people with a bit of automation exploring these light loops. ”
Seager and Pankratius also co-taught a course that focused on various aspects of the development of computing and artificial intelligence (AI) in planetary science. Seager says motivation for the course arose from a growing interest from students learning about AI and its applications to advanced data science.
In 2018, the course allowed students to use real data collected by TESS to study machine learning applications for this data. Modeled after another course taught by Seager and Pankratius, students of the course were able to select a scientific problem and learn the computer skills to solve that problem. In this case, students learned about AI techniques and applications to TESS. Seager says the students responded well to the special class.
“As a student, you could search,” Pankratius says. “You can build a machine learning algorithm, run on this data, and who knows, you might get something new.”
Much of the data TESS collects is also readily available as part of a larger citizen science project. Pankratius says that anyone with the right tools could start finding their own. Thanks to a cloud connection, this is even possible on a cell phone.
“If you’re tired of your bus journey home, why not find planets?” he says.
Pankratius says that this type of collaborative work allows experts in all fields to share their knowledge and learn from each other, rather than each trying to get caught up in a field the other.
“Over time, science has become more specialized, so we need ways to better engage the experts,” Pankratius says. The college of computing could create more such collaborations, he said. Pankratius also says it could attract researchers working at the crossroads of these subjects, who can close gaps in understanding between experts.
This type of work that integrates computer science is becoming increasingly common across scientific fields, Seager notes. “Machine learning is ‘in vogue’ right now,” she says.
Pankratius argues that this is partly because there is more evidence that the use of computer science methods is an effective way of dealing with various types of problems and a growing set of data sets.
“We now have demonstrations in a variety of areas where computer-aided detection just doesn’t work,” says Pankratius. “It will definitely lead to new discoveries. ”
https://ufc260livefreetv.eventsmart.com/2021/03/27/ufc-260-online-live-stream/
https://ufc260livefreetv.eventsmart.com/events/ufc-260-live-stream/
https://ufc260livefreetv.eventsmart.com/events/stipe-miocic-vs-francis-ngannou-ii-live-stream-reddit27-03-21/
https://ufc260livefreetv.eventsmart.com/2021/03/27/official-live-stipe-miocic-vs-francis-ngannou-ii-live-stream-reddit27-03-21/
https://vintage-camera-lenses.com/advert/watch-hd-belarus-vs-estonia-live-stream-online-free-tv-coverage-2021/
https://bisouv.com/