Communications of the ACM
Artificial Intelligence Analyzes Gravitational Lenses 10 Million Times Faster
A team at the Kavli Institute for Particle Astrophysics and Cosmology used artificial neural networks to analyze complex distortions in spacetime, demonstrating the method is 10 million times faster than traditional analyses.
Credit: Greg Stewart/SLAC National Accelerator Laboratory
Researchers at the U.S. Department of Energy's SLAC National Accelerator Laboratory and Stanford University have demonstrated that neural networks can accurately analyze gravitational lenses 10 million times faster than traditional methods.
The researchers note this type of analysis has been a time-consuming process that involves comparing images of lenses with a large number of computer simulations of mathematical lensing models, which can take months to complete. However, the neural networks enabled the researchers to complete the same analysis in only a few seconds.
The researchers trained the neural networks on about 500,000 simulated images of gravitational lenses.
"The neural networks we tested--three publicly available neural nets and one that we developed ourselves--were able to determine the properties of each lens, including how its mass was distributed and how much it magnified the image of the background galaxy," says the the U.S. National Aeronautics and Space Administration's Yashar Hezaveh.
From SLAC News Center
View Full Article
Abstracts Copyright © 2017 Information Inc., Bethesda, Maryland, USA
No entries found