In a post entitled Quantum Computing: Is the Future Here?, I noted that Lockheed Martin had purchased a quantum computer from the Canadian firm D-Wave. Now Google and NASA have teamed to purchase another D-Wave machine that will be the centerpiece of a new Quantum Artificial Intelligence Lab hosted at NASA’s Ames Research Center. [“Google, NASA Open New Lab to Kick Tires on Quantum Computer,” by Robert McMillan, Wired, 15 May 2013] According to McMillan, the computer is a D-Wave Two, the same model purchased by Lockheed Martin. On its corporate website, D-Wave describes their latest offering this way:
“The D-Wave TwoTM system is a high performance computing system designed for industrial problems encountered by Fortune 500 companies, government and academia. Our latest superconducting 512-qubit processor chip is housed inside a cryogenics system within a 10 square meter shielded room.”
The D-Wave Two is four times as powerful as its first quantum computer which housed a 128-qubit chip. Katherine Foley reports that the Google/NASA computer cost $15 million and is capable of producing “unheard-of calculation speeds 3600 times faster than those of conventional computers.” [“NASA Google Quantum Computer: The World’s Most Expensive Computer Thinks Like a Human,” Policymic, June 2013] Foley continues:
“The Canadian D-Wave-Two is the first commercially available computational system that supposedly utilizes quantum tunneling to solve complex mathematical equations. This process represents a complete overhaul of the way computer scientists have thought about processing.”
Foley’s use of the modifier “supposedly” reflects the fact the D-Wave system still has its detractors. McMillan explains:
“D-Wave has … met some skepticism from the quantum computing community. In part, it’s because D-Wave is taking a different approach to quantum computing. But it’s also because it hasn’t produced the kind of peer reviewed research on its systems that academics require.”
Clearly, D-Wave is confident enough in its system that it has been able to convince several large organizations, whose ranks are filled with some pretty smart people, that their quantum computer works. McMillan writes, “Although it’s still in the early days of experimentation, quantum computing could herald a new era of number-crunching.” He explains:
“That’s because it uses quantum physics to break computer processing out of the binary computing paradigm that has dominated for the past half-century. Instead of binary bits, these computers measure qubits, which can simultaneously represent many more values.”
In a blog post announcing the establishment of the Quantum Artificial Intelligence Lab, Hartmut Neven, Google’s Director of Engineering, wrote that the company’s interest in obtaining the D-Wave Two was to help it advance its research in machine learning. [“Launching the Quantum Artificial Intelligence Lab,” Research Blog, 16 May 2013] He wrote:
“We believe quantum computing may help solve some of the most challenging computer science problems, particularly in machine learning. Machine learning is all about building better models of the world to make more accurate predictions. If we want to cure diseases, we need better models of how they develop. If we want to create effective environmental policies, we need better models of what’s happening to our climate. And if we want to build a more useful search engine, we need to better understand spoken questions and what’s on the web so you get the best answer. … Machine learning is highly difficult. It’s what mathematicians call an ‘NP-hard’ problem. That’s because building a good model is really a creative act. As an analogy, consider what it takes to architect a house. You’re balancing lots of constraints — budget, usage requirements, space limitations, etc. — but still trying to create the most beautiful house you can. A creative architect will find a great solution. Mathematically speaking the architect is solving an optimization problem and creativity can be thought of as the ability to come up with a good solution given an objective and constraints. Classical computers aren’t well suited to these types of creative problems. Solving such problems can be imagined as trying to find the lowest point on a surface covered in hills and valleys. Classical computing might use what’s called ‘gradient descent’: start at a random spot on the surface, look around for a lower spot to walk down to, and repeat until you can’t walk downhill anymore. But all too often that gets you stuck in a ‘local minimum’ — a valley that isn’t the very lowest point on the surface. That’s where quantum computing comes in. It lets you cheat a little, giving you some chance to ‘tunnel’ through a ridge to see if there’s a lower valley hidden beyond it. This gives you a much better shot at finding the true lowest point — the optimal solution.”
Neven indicates that some of the theoretical mathematics has already been developed by Google and it’s been waiting for a computer powerful enough to execute the calculations. He explains:
“We’ve already developed some quantum machine learning algorithms. One produces very compact, efficient recognizers — very useful when you’re short on power, as on a mobile device. Another can handle highly polluted training data, where a high percentage of the examples are mislabeled, as they often are in the real world. And we’ve learned some useful principles: e.g., you get the best results not with pure quantum computing, but by mixing quantum and classical computing. Can we move these ideas from theory to practice, building real solutions on quantum hardware? Answering this question is what the Quantum Artificial Intelligence Lab is for. We hope it helps researchers construct more efficient and more accurate models for everything from speech recognition, to web search, to protein folding. We actually think quantum machine learning may provide the most creative problem-solving process under the known laws of physics.”
McMillan states, “The trick is getting these systems far enough along to solve real-world problems.” He goes on to report, however, that there is good news regarding the D-Wave Two.
“Researchers at Simon Fraser University and Amherst College presented a paper studying the D-Wave chip’s performance. They found that it worked pretty well on certain computing tasks.”
In more good news, Cade Metz reports that “researchers at the University of Southern California published a paper that comes … much closer to showing the D-Wave is indeed a quantum computer.” [“Google’s Quantum Computer Proven To Be Real Thing (Almost),” Wired, 28 June 2013] Metz continues:
“When those in the scientific community hear the term [quantum computer], they tend to think of a ‘universal quantum computer,’ a quantum computer that can handle any task. The D-Wave doesn’t work that way — it’s geared to particular calculations — but according to [Daniel Lidar, a professor of electrical engineering, chemistry, and physics at USC], the concepts behind it could be used, in theory, to build a universal quantum computer. Whatever you call it, the D-Wave is useful, helping to solve what are known as combinatorial optimization problems, which turn up in everything from genome sequence analysis and protein folding to risk analysis.”
The real value of the D-Wave computer will only be proven once results start coming in. Clearly, Lockheed, Google, and NASA believe there will be results and fairly soon.
