Close

Google Hires Quantum Computing Expert John Martinis to Build New Hardware

New team of rivals includes D-Wave and researchers that once cast aspersions on its quantum computer claims

2 min read
Google Hires Quantum Computing Expert John Martinis to Build New Hardware
Photo: Spencer Bruttig

Google recently unveiled its intention to build new quantum computing hardware—possibly laying the foundation for a super-fast computer capable of solving problems that would take forever on today's classical computers. The technology giant has already experimented with machines created by D-Wave, a Canadian company that says it has built the world's first commercial quantum computers. But Google's new project has recruited an outside expert with a very different mindset for transforming quantum computing into a practical technology.

Among the new faces at Google's quantum computing lab are John Martinis, a professor of physics at the University of California, Santa Barbara, and his quantum computing team. The Martinis group has been developing hardware based on superconducting quantum circuits. This is one of several ways to realize the potential of quantum computers, which are designed to perform many simultaneous calculations by taking advantage of quantum bits', or qubits', ability to exist as both ones and zeros at the same time. Google hopes to combine the Martinis team's expertise with lessons learned from tinkering with D-Wave's machines to design new quantum information processors based on a superconducting electronics approach.

"With an integrated hardware group, the Quantum AI team will now be able to implement and test new designs for quantum optimization and inference processors based on recent theoretical insights as well as our learnings from the D-Wave quantum annealing architecture," said Hartmut Neven, director of engineering at Google's Quantum Artificial Intelligence Lab, in a blog post.

A quantum computing coalition that includes D-Wave and academic researchers such as the Martinis group would have seemed almost unthinkable just several years ago. When D-Wave first made its confident claim of having built the "world's first commercially available quantum computer" in May 2011, many independent researchers were skeptical. They pointed out that D-Wave's unprecedented speed in scaling up to the 128-qubit D-Wave One machine seemed to ignore the issue of quantum coherence—the challenge of preserving the fragile quantum states of so many qubits. In other words, D-Wave's approach seemed to fly in the face of what physics said was practical.

"I always thought this was a gamble," Martinis told IEEE Spectrum previously. "Like other physicists, I thought it wouldn't work. But I admire them for going ahead and trying it."

Martinis still gave D-Wave credit for building an "amazing piece of technology" with a fairly sophisticated array of "superconducting electronics and control electronics." But Martinis's group has taken a slow but steady approach to building quantum computing hardware by experimenting with error correction codes capable of preserving fragile qubits long enough to perform calculations.

D-Wave rubbed many researchers the wrong way early on with its combination of uncompromising confidence and the relative secrecy surrounding the inner workings of its machines. But D-Wave has recently enjoyed a friendlier relationship with the research community by giving independent experts access to its machines leased by Lockheed Martin and a Google-NASA coalition. That change has probably helped set the stage for Google's umbrella effort to bring together former quantum computing rivals in the name of developing new hardware.

Martinis may have unintentionally foreshadowed his involvement with Google's quantum computing effort during a previous interview with IEEE Spectrum in the summer of 2013. At the time, he wrapped up the interview by saying he was glad that Google was demonstrating an interest in quantum computing by leasing the 512-qubit D-Wave Two machine.

"I hope that if they get to a point where they're looking for other options, they'll come talk to me and other people in the quantum computing community," Martinis said.

The Conversation (0)

The Future of Deep Learning Is Photonic

Computing with light could slash the energy needs of neural networks

10 min read

This computer rendering depicts the pattern on a photonic chip that the author and his colleagues have devised for performing neural-network calculations using light.

Alexander Sludds
DarkBlue1

Think of the many tasks to which computers are being applied that in the not-so-distant past required human intuition. Computers routinely identify objects in images, transcribe speech, translate between languages, diagnose medical conditions, play complex games, and drive cars.

The technique that has empowered these stunning developments is called deep learning, a term that refers to mathematical models known as artificial neural networks. Deep learning is a subfield of machine learning, a branch of computer science based on fitting complex models to data.

Keep Reading ↓ Show less