FileHippo News

The latest software and tech news

A machine powered by Artificial Intelligence, AI, has successfully managed to replicate and run a complicated Nobel prize winning experiment, all by itself. The... Experimental AI Successfully Replaces Physicists In Experiment

A machine powered by Artificial Intelligence, AI, has successfully managed to replicate and run a complicated Nobel prize winning experiment, all by itself.

The experiment, ironically developed by physicists working at the Australian National University, ANU, was an attempt to see how well, an AI controlled machine could be made responsible for managing the environmental conditions of a Bose-Einstein condensate; – essentially an extremely cold gas trapped in a laser beam.

AI

Typically speaking, Bose-Einstein condensates are themselves recreations of some of the coldest places in the universe, and for the experiment to run properly, had to be kept at less than one billionth of a degree above absolute zero.

What most surprised lead researcher on the project, Paul Wigley, was the speed and competency with which the AI got going:

“I didn’t expect the machine could learn to do the experiment itself, from scratch, in under an hour… [By contrast] …A simple computer program would have taken longer than the age of the Universe to run through all the combinations and work this out.”

In real terms, after the research team, built the machine and cooled the gas down to 1 micro kelvin, they gave control of 3 laser beams to the AI, sat back, watched, and were amazed at the results:

“It did things a person wouldn’t guess, such as changing one laser’s power up and down, and compensating with another…[coming] up with complicated ways humans haven’t thought of to get experiments colder and make measurements more precise.

The result of the AI test could change the way advanced experiments from all of science are done, the research team said, and could also lead to bigger and better experiments.

The artificial intelligence system’s ability to set itself up quickly every morning and compensate for fluctuations such as changes in the Earth’s magnetic field could make the currently lab bound technology much more useful for field measurements.

“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 researcher, Dr Michael Hush.

“It’s cheaper than taking a physicist everywhere with you.”

The research is published in the Nature group journal Scientific Reports.