Quantum computing helps refine Google Glass


Google has revealed that its upcoming wearable computer, Glass, will in part be powered by quantum computing – although what that actually means remains unclear.

Earlier this year Google announced that it had purchased a quantum device from D-Wave of Burnaby, Canada, and partnered with NASA to set up the Quantum Artificial Intelligence Lab. In a video that debuted today at the Imagine Science Films Festival at Google New York, the team discuss the types of applications they will explore with the D-Wave computer.

Quantum computers process information in quantum bits, or qubits, which can be both 0 and 1 at the same time. In theory, this should one day allow quantum chips to vastly outperform regular PCs. D-Wave’s computers remain controversial, because they use a non-mainstream technique called adiabatic computing that does not conclusively exploit quantum phenomena to increase performance.

But a paper released earlier this year demonstrated that D-Wave could beat PCs running general-purpose optimisation software, which is used in applications from image recognition to machine learning. The new Google video says that the team wants to use D-Wave’s abilities for tasks such as analysing climate or genomics data and determining whether we are alone in the universe.

“The overwhelming obvious killer app for quantum computation is optimisation,” says D-Wave founder Geordie Rose in the Google video.

Better blinking

The first real-world application to come out of the lab may be improved algorithms for Google Glass. The headset itself is not able to run quantum software, but it is possible that a quantum computer could optimise an ordinary algorithm for use on the low-powered wearable device.

Although the project is not specifically mentioned in the video, tech site The Verge reports that Google engineers have already used D-Wave to design a better blink-detection algorithm, which will allow Glass users to “click” on links by blinking. That fits with an earlier blog post by Rose, which also said that Google had used D-Wave to train their blink detectors.

So far, though, D-Wave has not published any results showing that its machines can carry out optimisation problems better than traditional devices that are specifically programmed to solve a given problem. Without this evidence, Google’s new video has left quantum computing experts unconvinced.

“While I don’t know the details of this Google Glass demo, I’m sceptical that there’s anything here that couldn’t be done much faster and more easily using classical computers,” says Scott Aaronson of the Massachusetts Institute of Technology.

Best questions

Matthias Troyer of the Swiss Federal Institute of Technology (ETH) in Zurich has previously studied the performance of D-Wave’s computers, and is also unsure of the milestone. “This was probably a technology demonstration, showing that such a kind of problem can, in principle, be solved on a D-Wave machine. However, any problem that the machine might have solved could easily have been done on a PC.”

Even if D-Wave’s computers do not outperform regular machines, they are still worthy of study, says Troyer. “Just building a device that can do what the D-Wave machine does is quite an engineering achievement.”

Google and D-Wave did not respond to requests for comment before this article was published. But the collaboration emphasises in its video that they are still in the early days of exploring the technology. “We don’t know what the best questions are to ask that computer – that’s exactly what we’re trying to understand now,” says Eleanor Rieffel of the NASA Ames Research Center in Mountain View, California.


About Author

%d bloggers like this: