backtop


Print 20 comment(s) - last by random2.. on Oct 11 at 6:16 PM

Canadian startup receives funding from two key sources

Situated in the verdant North American northwest, Canada's University of British Columbia proved the fertile breeding ground for the first major "quantum computing" startup.  Calling itself "D-Wave", the company launched in 1999 with an odd assortment of professors and graduate students.  One of those students -- who at the time was just graduating with his Ph.D -- was Geordie Rose.  Today Mr. Rose, one of the company co-founders, is the company's chief technology officer (CTO) and is fighting a battle to convince venture capitalists (VCs) that his company is dealing in reality, not peddling digital snake oil.

I. Making Quantum Computing a Cold, Hard Reality

Like the minicomputers of decades past, D-Wave's systems -- first unveiled in 2007 -- are hulking designs, 10 feet high.  But the majority of that space is occupied by the supercooled gases and cooling "gun", which chills a custom chip down.  

D-Wave quantum computing
D-Wave's "Quantum" computers [Image Source: D-Wave Systems]

According to the Massachusetts Institute of Technology's publication Technology Review, the top of the gun is a balmy -269 °C (4 Kelvin) -- just slight warmer than the background radiative temperature in the chilly depths of space (2.7 Kelvin).  At its tip, the gun cools a 1-inch square chip to millikelvin temperatures -- colder than outer space, colder even than Large Hadron Collider's operating temperature of 1.9 Kelvin.

D-Wave cooling gun
D-Wave's cooling gun chills the company's chips to colder than space temperatures.
[Image Source: ZDNet]

While researchers today are probing using nanostructures to store encode information in electron spin -- forming so-called single atom "quBits" (quantum bits) -- the D-Wave chip is a cruder design, consisting of niobium alloy loops that trap current either clockwise (0) or counter-clockwise (1).  Superconductors called couplers selectively link the "quBit" loops, allowing them to interact, and flip each other.

D-Wave chips
D-Wave's 16 quBit "Orion" quantum processor. [Image Source: D-Wave Systems]

The system requires two forms of guidance.  First it needs a quantum algorithm -- which consists of setting operating settings of the couplers.  Second, it needs raw data for that algorithm to churn on.

While creating the quantum algorithm and converting data into a digestible format for the device is a daunting task, the potential gains are great.  The system can quickly equilibrate, with the quBits quickly coalescing on a solution to an optimization problem -- represented by a lower-energetic configuration of the system.  

The optimization process involves the quBits becoming entangled -- during which time they can occupy both 0 and 1 simultaneously.  (This claim has been challenged as unproven by critics, who suggest the system could be operating with traditional nanoscale electro-magnetic effects.)

Comments Geordie Rose, Ph.D, "Virtually everything has to do with optimization, and it's the bedrock of machine learning, which underlies virtually all the wealth creation on the Internet."

II. Critics Question Whether D-Wave Computer is Quantum or Simply "Weird"

Despite the fact that the company was spawned out of a leader in the academic world's quantum computing push, skepticism from the research community has been long simmering on at least three grounds.  

Quantum computer
Skepticism over D-Wave quantum claims has raged. [Image Source: Physics World]

First, there's a war over terminology -- today's theoretical quantum computers share some elements (such as reliance on entanglement) with D-Wave's quBits, but physically are very different.  That leads to the second criticism -- that D-Wave's empirical development process has created a situation in which they don't know exactly what is going on inside their chip, and thus cannot properly describe potential differences with alternative designs  (e.g single atom quBit).  Lastly, many complain that D-Wave's algorithms aren't yet even as fast as classical transistor-based logic.

MIT Professor Scott Aaronson -- a quantum-computing expert and long-time critic -- comments, "At an engineering level they've put together a setup that's impressive in various ways.  But in terms of the evidence that they're solving problems using quantum mechanics faster than you could classically, I don't think it's there yet."

Professor Aaronson and others are concerned that despite support from experts at certain institutions -- such as The National Aeronautics and Space Administration (NASA) -- D-Wave has simply built a "weird" computer, not a quantum computer at all.  After D-Wave researchers published in Nature a paper last year describing quantum annealing, it became clear that the D-Wave chips were demonstrating some quantum effects.  Professor Aaronson blogged on the development, slightly softening his criticism.

Albert Einstein
D-Wave has yet to prove its systems operating using quantum entanglement, a key feature of quantum computers, which Albert Einstein described as "spooky". [Image Source: Newscom]

But he argues D-Wave still needs to demonstrate definitive proof of two things -- first evidence that quantum entanglement is occurring on-chip, and second evidence definitively demonstrating a "quantum speedup" over classical designs.  And even if D-Wave can show those things, says Professor Aaronson, its "hard coded" algorithm approach and requirement of heavily pre-processed data make it a less than ideal quantum computer solution.

III. High Profile VC Backers Boost Canadian Company

But D-Wave is forging ahead.  It recently raised $30M USD in a new wave of funding.  Among the high-profile donors were Amazon.com, Inc. (AMZN) founder and CEO Jeff Bezos and U.S. Central Intelligence Agency's (CIA) venture capital branch, In-Q-Tel.  For Mr. Bezos, the interest is likely applying quantum chips in search and other internet applications, while the CIA's interest most likely focuses on code-breaking/encryption.

Jeff Bezos
Amazon founder and tech visionary Jeff Bezos recently became a D-Wave investor.
[Image Source: Getty Images]

D-Wave already has a growing stable of high-profile clients.  Since 2011 Google Inc. (GOOG) researchers have been paying D-Wave for access to the system, remotely logging in to D-Wave's control systems to test out quantum image search and other novel algorithms.  

And Lockheed Martin Corp. (LMT), one of the world's largest defense contractors, paid D-Wave $10M USD for one of the D-Wave systems, hoping to use the quantum algorithms to track down bugs in the mission-critical code-base for the delayed F-35 fighter jet project.


Lockheed is using the quantum compute to hunt for bugs in its F-35B Lightning II fighter code
[Source: Lockheed Martin]

The high profile support shows that a lot of smart people with a lot of resources believe in D-Wave.  After all, for all the questions, for all the delays in upping the quBit count as promised, D-Wave is one of a kind.  It is the world's first firm to claim to offer a "commercially available quantum computer."

Sources: D-Wave [press release], Nature, Scott Aaronson, MIT, Technology Review



Comments     Threshold


This article is over a month old, voting and posting comments is disabled

RE: In Quantum Mechanics.
By ARoyalF on 10/7/2012 10:43:16 PM , Rating: 2
Hello Werner!


"And boy have we patented it!" -- Steve Jobs, Macworld 2007














botimage
Copyright 2014 DailyTech LLC. - RSS Feed | Advertise | About Us | Ethics | FAQ | Terms, Conditions & Privacy Information | Kristopher Kubicki