John Martinis is one of the worlds foremost experts on quantum computing, a growing field of science that aims to process information at super high speeds using strange physics of very tiny particles such as electrons and photons. And now, after years as a physics professor at the University of California Santa Barbara, hes headed for Google.

This week, the Google Quantum A.I. Lab announced that it hired Martinis and his Santa Barbara team to build a new breed of quantum computing hardware. Though Martinis will maintain his affiliation with UC Santa Barbara and continue to mentor his PhD students there, he will spend most of his time on his research at Google. The move proves that Google is serious about quantum computing, and given the companys vast influence and deep pockets, it could provide a serious shot in the arm for quantum computer research as a whole.

Google launched its Quantum A.I. Lab last year to test a machine called the D-Wave Two, an intriguing but controversial system that its makers bill as a quantum computer, and it believes quantum computing could play a key role in so many of its future ambitions, from self-driving cars and other robots to better predictive analytics systems for products like Google Now to things we havent even dreamed up yet. Thanks to whats called the superposition principle of quantum mechanics, it could process data for such projects at speeds that are exponentially faster than what you get from todays machines.

But the scientific community has greeted the D-Wave machine with skepticism, questioning whether the machine is actually a quantum computer at all, and whether it can actually provide something you cant get from conventional machines. In joining Google, Martinis lends new weight to the companys quantum ambitions.

Martinis is among those questioning D-Waves claims. Last June, Science published a paper co-authored by Martinis and several other scientists concluding that D-Waves machines arent actually faster than normal laptops and desktops. But hes no D-Wave hater. Martinis has been working with D-Waves machines for a few years now and says he has long been impressed with the work the company has done.

The general consensus now, he says, is that the D-Wave computers do exhibit some quantum behavior. The real question, he explains, is whether this behavior actually speeds up the D-Wave computers. And although his team will be working separately from D-Wave at Google, he thinks their work may eventually help D-Wave take better of advantage of that quantum behavior. Were taking some of the basic ideas of D-Wave and combining that with what the [Google] Quantum AI team has learned operating the machine, he says.

Martinis and his team have been focused on stability of quantum computers, one of the biggest issues in this field of research. Quantum computers rely on particles that are in whats called a quantum state, meaning that they can be two different states at once, or even some combination of the two. The problem is that these particles, called qubits, typically stay in a quantum state for only a fraction of a second, and interference from other particles can easily be knocked out of this state.

Researchers call this loss of quantum information decoherence, and its made it hard to make any reliable calculations with a quantum computer. But Martinis and company are making some progress in this area. Last April, Nature published a paper detailing their work creating a five qubit computer that can maintain quantum information using superconductive materials and an error correction system that creates back-up qubits for each piece of information store. Now, theyve managed to scale it up into a nine qubit machine, and hope to double the number of qubits each year.

Meanwhile, D-Wave has been mostly focused on trying to build machines with as many qubits as possible, but it hasnt focused much on the problem of decoherence, Martinis says. By combining D-Waves work on achieving scale with their own work on stability, Martinis and his team think they can push the whole field of quantum computing further.

Martinis says that joining Google has a few distinct advantages. One is that the Google Quantum A.I. Lab Team team has developed some real-world applications for quantum computing, so Martinis and company will have some real problemssuch as complex route planningto sink their teeth into. Another is that it will enable him to build a full-time, permanent team, as opposed to a staff of grad students and post-doctoral students who come and go every couple years.

See the original post here:

The Man Who Will Build Googles Elusive Quantum Computer