Quantum randomness of empty space can be controlled with a laser

The minuscule fluctuations of seemingly empty space can be controlled just enough to make the building blocks of a new type of computer.

Empty space is not actually empty – it is filled with tiny flickers of quantum fields. Boosting those flickers with very weak laser light can turn the apparent nothingness into a building block for a new kind of light-based computer.

The experimental setup passes light through lenses, mirrors and a crystal
Charles Roques-Carmes, Yannick Salamin

“Suppose you had a completely empty, dark box and then you put an electric field detector inside the box. If you averaged all its measurements you would get zero, but each individual measurement would be a little more or a little less than zero,” says Charles Roques-Carmes at Stanford University in California. He and his colleagues have now worked out how to control those “vacuum fluctuations” with enough accuracy to store information.

The researchers built a device consisting of an intricate maze of components, such as lenses and mirrors, that bounce and shape light. The most important of these is a crystal. When hit with a laser, it emits pulses of light that have a property called phase, which can be interpreted as one of two numbers: 1 and 0. The number it lands on depends on the vacuum fluctuations.

Vacuum fluctuations assume values below and above zero randomly, so they change the phase number randomly as well. This creates unpredictable sequences of 1s and 0s, which researchers have previously used to generate random numbers. But Roques-Carmes and his colleagues biased that randomness by adding a second, much weaker laser beam to their setup, which could give a small boost to only some vacuum fluctuation and tip the scale towards more 1s or more 0s.

In this way, they created a “probabilistic bit”, or p-bit, where 1s occurred reliably with one probability and 0s with another, different probability. They could then use these p-bits to store basic information. This is a first step towards building a probabilistic computer, or one that can give a range of answers when it repeats a calculation instead of always returning the same result.

Probabilistic computers could be useful because many problems naturally involve probabilities, including many optimisation and some quantum simulation problems, says Kerem Çamsari at the University of California, Santa Barbara.

Supriyo Datta at Purdue University in Indiana, who previously created p-bits out of unstable magnets in collaboration with Çamsari and other colleagues, says that making them out of light may lead to innovation in how multiple p-bits can be connected to make a bigger and more powerful probabilistic computer.

The approach is simpler than other unconventional computing methods, such as those involved in quantum computing that require extremely low temperatures, says Alfred Leitenstorfer at the University of Konstanz in Germany. The new experiment spans a few metres of space and creates a single p-bit, so it would have to be miniaturised to be truly convenient, he says.

“I was also thrilled to see this work because I think there are still physicists around who think that these vacuum fluctuations are something that you just need in theory. And this work demonstrates that, no, they are real quantities,” he says.

Journal reference:

Post a Comment