Machine Learning Slashes Tech Design Process by a Whole Year
Imagine life moving at 40,000 times the current speed. A flight from New York to Los Angeles would take a mere half a second, and a tomato would be ripe three minutes after its seed was planted.
A research team at the Sandia National Laboratories (Sandia) in the U.S. has found a way to improve machine learning so that the design process of materials for new technologies could be 40,000 times faster.
Their research was published in Computational Materials on Monday.
The team at Sandia managed to use machine learning to complete materials science calculations at 40,000 times the regular speed.
This new advancement could change the way new technologies for optics, aerospace, energy storage, and even medicine are made. On top of this, the laboratories creating these technologies could save money on computing costs.
"We’re shortening the design cycle," explained David Montes de Oca Zapiain, a computational materials scientist at Sandia who helped lead the research.
"The design of components grossly outpaces the design of the materials you need to build them. We want to change that. Once you design a component, we’d like to be able to design a compatible material for that component without needing to wait for years, as it happens with the current process."
How fast the machine learning algorithm operates
As an example, the team noted that a single, unassisted simulation on a high-performance computing cluster of 128 processing cores took 12 minutes. With their new machine learning, however, this exact same simulation took just 60 milliseconds using only 36 processing cores equivalents. That's 42,000 times faster than without machine learning.
In other words, what would typically take a year for researchers to learn now takes only 15 minutes.
"Our machine-learning framework achieves essentially the same accuracy as the high-fidelity model but at a fraction of the computational cost," said Sandia materials scientist Rémi Dingreville, who was also part of the project.
The team is looking to the future and its next steps see it using its algorithm to research ultrathin optical technologies to be used on monitors and screens.
This summer saw a surge in travelers, and their lost or mishandled luggage. Can AI save the mounting baggage crisis?