Researchers apply machine learning to condensed matter physics

February 13, 2017, Perimeter Institute for Theoretical Physics

A machine learning algorithm designed to teach computers how to recognize photos, speech patterns, and hand-written digits has now been applied to a vastly different set of data: identifying phase transitions between states of matter.

This new research, published today in Nature Physics by two Perimeter Institute researchers, was built on a simple question: could industry-standard algorithms help fuel physics research? To find out, former Perimeter Institute postdoctoral fellow Juan Cassasquilla and Roger Melko, an Associate Faculty member at Perimeter and Associate Professor at the University of Waterloo, repurposed Google's TensorFlow, an open-source software library for machine learning, and applied it to a physical system.

Melko says they didn't know what to expect. "I thought it was a long shot," he admits.

Using gigabytes of data representing different state configurations created using simulation software on supercomputers, Carrasquilla and Melko created a large collection of "images" to introduce into the machine learning algorithm (also known as a neural network). The result: the distinguished phases of a simple magnet, and could distinguish an ordered ferromagnetic phase from a disordered high-temperature phase. It could even find the boundary (or transition) between phases, says Carrasquilla, who now works at quantum computing company D-Wave Systems.

"Once we saw that they worked, then we knew they were going to be useful for many related problems. All of a sudden, the sky's the limit," Melko says. "Everyone like me who has access to massive amounts of data can try these standard neural networks."

This research, which was originally published as a preprint on the arXiv in May, 2016, shows that applying machine learning to condensed matter and statistical physics could open entirely new opportunities for research and, eventually, real-world application.

Explore further: Success by deception

More information: Machine learning phases of matter, Nature Physics, nature.com/articles/doi:10.1038/nphys4035

Related Stories

Success by deception

February 13, 2017

Theoretical physicists from ETH Zurich deliberately misled intelligent machines, and thus refined the process of machine learning. They created a new method that allows computers to categorize data—even when humans have ...

The thermodynamics of learning

February 6, 2017

(Phys.org)—While investigating how efficiently the brain can learn new information, physicists have found that, at the neuronal level, learning efficiency is ultimately limited by the laws of thermodynamics—the same principles ...

Recommended for you

On the rebound

January 22, 2018

Our bodies have a remarkable ability to heal from broken ankles or dislocated wrists. Now, a new study has shown that some nanoparticles can also "self-heal" after experiencing intense strain, once that strain is removed.

Nanoparticle gel controls twisted light with magnetism

January 22, 2018

"Help me, Obi Wan Kenobi. You're my only hope." For many of those around at the release of Star Wars in 1977, that scene was a first introduction to holograms—a real technology that had been around for roughly 15 years.

1 comment

Adjust slider to filter visible comments by rank

Display comments: newest first

rderkis
5 / 5 (1) Feb 13, 2017
Simulations Rule!
And Quantum computer simulations will change everything.

Please sign in to add a comment. Registration is free, and takes less than a minute. Read more

Click here to reset your password.
Sign in to get notified via email when new comments are made.