The technology could help us “work our way towards uncovering the greatest mystery of modern physics, which is to uncover what dark energy is”, according to the lead author of a study detailing the breakthrough.
Astronomers have long been searching for a full picture of the universe. That has brought major breakthroughs as researchers have looked more deeply and in more detail at the universe than ever before.
But researchers continue to run up against the problem of dark matter and dark energy, which makes up most of the universe but can only be observed indirectly. Nobody has been able to understand its nature or how it fits into any possible model of the cosmos.
Some scientists have hoped to move towards solving this problem with a greater understanding of the structure of the universe. If astronomers are able to understand how our cosmos is made up – generating a full picture of the filaments where galaxies cluster together, and the voids that appear to be empty – then they may be able to give a better picture of what it is made of, and how it works.
Astronomers have been able to make some progress towards a fuller picture of that with the discovery of the cosmic microwave background, which has allowed them to build a picture of the universe as it looked very soon after its beginning.
But they hope to build a deeper understanding of the universe got from that very early picture to the one that surrounds us today. If they were able to build a picture of that journey, it could help iluminate the mystery of dark matter and dark energy.
Now a team of scientists has built a new tool called the “Dark Emulator”, described in a recently published paper in the Astrophysical Journal. They hope it could use artificial intelligence to shed more light on how that structure came about.
The technology was built by scientists who used supercomputers belonging to the National Astronomical Observatory of Japan. They were able to use that emulator on data taken from many of the world’s largest surveys of the sky, allowing them to examine how the cosmic structures of our universe could have been formed.
“We built an extraordinarily large database using a supercomputer, which took us three years to finish, but now we can recreate it on a laptop in a matter of seconds. I feel like there is great potential in data science,” said lead author Takahiro Nishimichi in a statement.
“Using this result, I hope we can work our way towards uncovering the greatest mystery of modern physics, which is to uncover what dark energy is. I also think this method we’ve developed will be useful in other fields such as natural sciences or social sciences.”
The tool works by relying on machine learning to create hundreds of virtual universes. They were able to alter important parts of the universe, and generate a picture of how they might have changed the cosmos.
The “Dark Emulator” is able to learn from each of those simulations, allowing it to guess how changes in those characteristics would affect the results. That means it does not need to create entirely new simulations, allowing it to generate another virtual universe far more quickly than ever before.
In the paper describing the findings, the researchers describe how it was able to accurately predict specific effects in their virtual universes in just seconds. Similar simulations would take days without using the Dark Emulator, they say.