Although physics might seem like a field in which human input will always be essential, this notion is now being challenged by researchers from the Australian National University (ANU) and the University of New South Wales at the Australian Defense Force Academy (UNSW ADFA) who have successfully used artificial intelligence to run a physics experiment.
The experiment involved creating a Bose-Einstein condensate - a very cold gas trapped in a laser beam - and is a replication of the experiment that nabbed the 2001 Nobel Prize.
"I didn't expect the machine could learn to do the experiment itself, from scratch, in under an hour," said Paul Wigley from the ANU Research School of Physics and Engineering and co-lead author of the study. "A simple computer program would have taken longer than the age of the universe to run through all the combinations and work this out."
Bose-Einstein condensates make up some of lowest temperatures in the universe and have a high degree of sensitivity to changes in external stimuli. This sensitivity gives them the potential to be utilized for taking precise measurements in fields such as mineral exploration and navigation systems.
Using an artificial intelligence system would help scientists integrate these condensates into such fields to take measurements and provide the benefit of being able set itself up and compensate for any fluctuations that occur overnight.
"You could make a working device to measure gravity that you could take in the back of a car, and the artificial intelligence would recalibrate and fix itself no matter what," said Michael Hush from UNSW ADFA and co-lead author along with Wigley. "It's cheaper than taking a physicist everywhere with you."
After cooling the gas to approximately one microkelvin, the team handed over three laser beams to the artificial intelligence, which then cooled down the gas trapped inside the laser beams to a few hundred nanokelvins.
The artificial intelligence not only completed the experiment, but it also surprised the team with the methods that it used to do so.
"It did things a person wouldn't guess, such as changing one laser's power up and down, and compensating with another," Wigley said. "It may be able to come up with complicated ways humans haven't thought of to get experiments colder and make measurements more precise."
"Next we plan to employ the artificial intelligence to build an even larger Bose-Einstein condensate faster than we've seen ever before," he concluded.
The findings were published in the May 16 issue of Scientific Reports.
