Photo by: NASA

NASA

Astronomy’s Newest Tool: Artificial Intelligence

In a bowl, combine a tablespoon of hydrogen and a teaspoon of helium to a cup of dark matter. Add a pinch of neutrinos and sprinkling of radiation. Mix well to combine. Heat to several million Kelvin. When mixture has risen, leave to cool for 13 billion years.

August 05, 2021

No, it’s not a recipe for your grandma’s famous chocolate chip cookies – it’s how the universe is made. We understand the fundamental ingredients of the cosmos (which is over 80% dark matter, a form of matter that does not interact with light, plus a decent amount of normal matter and a bit of other stuff), and we know how those ingredients interact (mostly through gravity).

We know that the early universe was far smaller, far hotter, and far denser than the cosmos of today. We also know that it was almost perfectly uniform, with only one-in-a-thousand differences in density from place to place.

We know that when the ingredients of the universe interact through gravity, those tiny differences grow up to become enormous differences. Small regions of slightly-denser-than-average clumps attract more matter on to them, eventually building up to become stars, groups, galaxies, and more.

How do we know that dark matter isn't just normal matter exhibiting strange gravity? A new observation of gravitationally magnified faint galaxies far in the distance behind a massive cluster of galaxies is shedding new dark on the subject. This image from the Hubble Space Telescope indicates that a huge ring of dark matter likely exists surrounding the center of CL0024+17 that has no normal matter counterpart.

Photo by: NASA, ESA, M. J. Jee and H. Ford et al. (Johns Hopkins Univ.)

NASA, ESA, M. J. Jee and H. Ford et al. (Johns Hopkins Univ.)

How do we know that dark matter isn't just normal matter exhibiting strange gravity? A new observation of gravitationally magnified faint galaxies far in the distance behind a massive cluster of galaxies is shedding new dark on the subject. This image from the Hubble Space Telescope indicates that a huge ring of dark matter likely exists surrounding the center of CL0024+17 that has no normal matter counterpart.

But while it’s easy to say, it’s harder to study. That’s because when all the ingredients start really getting together and building large structures in our universe, we can’t follow along with simple equations on the chalkboard. Instead, we have to turn to computer simulations to do science: we input into the simulations what we think the universe is made of and how those ingredients interact, we let the simulation evolve through billions of years of cosmic history, and we compare the simulated universe to real one to see how close we got.

These simulations can be especially time-consuming, especially when trying to recreate large sections of the universe to very fine detail.

Thankfully, we have some tricks up our sleeve to make them go a little faster. The trick is here is a kind of artificial intelligence known as machine learning. To make this work, researchers ran a bunch of low-resolution, easy-to-run simulations. They then let an artificial neural network find the best way to connect those low-resolution simulations to high-resolution ones. The neural network then compared the scaled-up simulations to versions made with traditional (i.e., expensive) methods.

With the links established, the researchers could then just run the low-resolution simulations and be able to make predictions that could be compared to the real universe, potentially radically speeding up our ability to test theories and make accurate models of the universe.

This is a portion of the Hubble Ultra Deep Field North image, which encompasses infrared, visible and ultraviolet wavelengths and shows thousands of galaxies. It includes very distant galaxies, which can only be seen in infrared light, and closer galaxies, which can be seen in wavelengths that include visible and ultraviolet light.

Photo by: NASA, ESA, P. Oesch (University of Geneva) and M. Montes (University of New South Wales)

NASA, ESA, P. Oesch (University of Geneva) and M. Montes (University of New South Wales)

This is a portion of the Hubble Ultra Deep Field North image, which encompasses infrared, visible and ultraviolet wavelengths and shows thousands of galaxies. It includes very distant galaxies, which can only be seen in infrared light, and closer galaxies, which can be seen in wavelengths that include visible and ultraviolet light.

Their work isn’t done, however. Right now their setup is relatively simple – it only includes dark matter. But real observations are based on the positions of the galaxies and stars that we can see with our telescopes. Big, fancy, slow simulations include models of star and galaxy formation, and so right now the only way to compare 1-to-1 with observations is through the traditional route.

The next step for the researchers is to include the formation of stars and galaxies into the low-resolution simulations, hoping that they can pull of the machine learning trick again with a more complicated setup.

Dive Deeper into the Cosmos

Journey Through the Cosmos in an All-New Season of How the Universe Works

The new season premieres March 24 on Science Channel and streams on discovery+.

Paul M. Sutter

Paul M. Sutter is an astrophysicist at Stony Brook University and the Flatiron Institute, host of Ask a Spaceman and Space Radio, and author of How to Die in Space.

Next Up

The Cost of Global Satellite Internet: Worse Astronomy

A few billionare-backed companies have ambitious goals: launching tens of thousands of communication satellites to provide global high-speed internet access. Elon Musk’s StarLink, Jeff Bezos’ Project Kuiper, One Web, GuoWang, and more are all competing for this lucrative market. In less than a decade, we can expect over 50,000 new satellites to encircle the Earth. That’s about ten times more than are currently active.

Who Wants to Be an Astronaut?

If you've ever wanted to travel into space, this is your chance. No, really. Even you.

Romeo and Juliet: The Story of Galaxy Collisions

Our Milky Way galaxy is on a collision course. With destiny. With destruction. With fate. With our nearest neighbor, Andromeda. You can stream HOW THE UNIVERSE WORKS on discovery+.

Michael Collins, Apollo 11 Astronaut, Dies at Age 90

After a truly legendary life, NASA astronaut Michael Collins passed away at the age of 90. Michael was a vital member of the Apollo 11 mission to the moon in 1969 and a pioneer of space exploration.

Want to See a Black Hole’s Magnetic Field? Now’s Your Chance

A couple years ago, the team of astronomers with the Event Horizon Telescope wowed the world by providing our first-ever snapshot of a real-life black hole. Now they’ve done one better and mapped out the swirling magnetic fields around the monster. It’s our first ever glimpse of the forces that power the largest engines in the universe.

What Screaming Black Holes are Telling Us

In 2002, NASA’s orbiting X-ray observatory, the Chandra telescope, mapped out the movements of hot gas in a cluster of galaxies sitting 250 million light-years away.

Do You Want to Go to Space?

Have you always dreamed of going to space? Former NASA astronaut Mike Massimino answers our questions about life at the International Space Station.

Astronomers See Flashes from Behind a Black Hole

Want to see what’s behind a black hole? Easy. You just…stare at it. The whole thing is pretty weird to contemplate, but an excellent example of the space-bending (and mind-bending) powers of black holes.

The Milky Way Broke its Arm (But is Totally Okay)

The Milky Way is a giant, magnificent, truly transcendently beautiful spiral arm galaxy. It’s too bad we can’t get a decent picture of it. The problem is that we live inside it, and so astronomers have to work extra-hard to construct an accurate map.

When We’ll Know if NASA’s Asteroid Impact Test was a Success

Recently NASA’s DART mission succeeded in its primary goal, which was to slam a spacecraft face-first into an asteroid. For science. The intention of the mission was to test if we could actually redirect an asteroid and send it into a different orbit. But how and when will we know if it worked?

Related To: