Facial recognition isn’t just for, well, faces: The machine learning method has shown promise in helping astronomers analyze images of galaxies.

Using mock Hubble Space Telescope snapshots, researchers trained a deep learning system to recognize key phases of galaxy evolution.

When shown a large set of actual Hubble photos to classify, the neural network showed a “remarkable” level of consistency.

“We were not expecting it to be all that successful. I’m amazed at how powerful this is,” University of California Santa Cruz professor Joel Primack said in a statement. “We know the simulations have limitations, so we don’t want to make too strong a claim. But we don’t think this is just a lucky fluke.”

Deep learning has widely been applied to facial ID and other image- and speech-recognition applications. But it is also surprisingly good at classifying real galaxies in photos.

Star systems are notoriously complex, changing appearance as they evolve; following the evolution of an individual galaxy over time is possible only in simulations.

In a new study, published recently by Astrophysical Journal, the researchers explored whether deep learning can detect a phenomenon known as “blue nugget.”

Turns out it can.

Described as the process of young, hot stars emitting short “blue” wavelengths, the BN phase was spotted by the computer program in simulated and observational data.

“It may be that in a certain size range, galaxies have just the right mass for this physical process to occur,” co-author David Koo, professor of astronomy and astrophysics at UC Santa Cruz, said.

The benefits of machine learning in astronomy are countless.

“Deep learning looks for patterns, and the machine can see patterns that are so complex that we humans don’t see them,” Koo continued. “We want to do a lot more testing of this approach, but in this proof-of-concept study, the machine seemed to successfully find in the data the different stages of galaxy evolution identified in the simulations.”

And this is just the start. As large survey projects and new telescopes contribute to a growing pile of observational data, deep learning methods will become powerful tools for interpretation.

“This is the beginning of a very exciting time for using advanced artificial intelligence in astronomy,” Koo added.

 

Let us know what you like about Geek by taking our survey.





READ SOURCE

LEAVE A REPLY

Please enter your comment!
Please enter your name here