that actually has had the bravado to claim
such a demonstration is Canadian firm D-Wave Systems, Inc. which
is based out of Barnaby, British Columbia. The company has
developed what it claims to be working 16-qubit, 28-qubit, and
128-qubit quantum computer chips. Each qubit is implemented
with a magnetically coupling superconducting loop called rf-squid
flux. The company has fabricated some of these chips at NASA's
Jet Propulsion Lab’s microdevices lab in Pasadena and NASA
scientists who saw the work firsthand back
its credibility, despite widespread
doubts in the research community.
Now D-Wave has received
an even bigger endorsement, from the world's largest internet firm:
Google. Google manager Hartmut Neven announced in a blog
post last week that his company had been working with D-Wave to
develop quantum computers to power a search of still images in a
database of images, video, and PDFs. The project has been
ongoing for three years according to Mr. Neven.
elaborating, "Over the past three years a team at Google has
studied how problems such as recognizing an object in an image or
learning to make an optimal decision based on example data can be
made amenable to solution by quantum algorithms. The algorithms we
employ are the quantum
adiabatic algorithms discovered by Edward
Farhi and collaborators at MIT. These algorithms promise to find
higher quality solutions for optimization problems than obtainable
with classical solvers."
The announcement corresponded
with the first demonstration of the fruits of the partnership.
At the Neural Information Processing Systems conference (NIPS
2009), Google showed off a working search that could locate
images of cars in the database almost instantly after being first
trained images of what a car looked like. The search training was
powered by D-Wave's new C4 Chimera chip and used the quantum
adiabatic image algorithms.
In the search, Google first took
20,000 photographs -- half with cars in them and without. In
each picture they drew boxes around the cars (if there were any),
identifying the "car" graphic element. Next Google
took a second set of 20,000 photos -- half with cars and half
without. They then put the second set to the quantum trained algorithm, which identified the cars faster than any traditional
algorithm in Google's data farms.
Google was quite
enthusiastic about the results. Writes Mr. Neven, "There
are still many open questions but in our experiments we observed that
this detector performs better than those we had trained using
classical solvers running on the computers we have in our data
centers today. Besides progress in engineering synthetic intelligence
we hope that improved mastery of quantum computing will also increase
our appreciation for the structure of reality as described by the
laws of quantum physics."
Could quantum computing be
Google's trump card to keep down a resurgent
Microsoft, which has been invigorated by its partnerships with
Yahoo and Wolfram
Alpha? If the technology is as good as Google claims, the
only real question seems to be how long it will take Google to make
its deployment affordable. D-Wave's past designs were complex
beasts that needed to be chilled to almost 0 degrees Kelvin to
operate properly. Still, Google and D-Wave have both come a long way
in terms of quantum hardware and software, so we may not have to wait
too long for quantum-computer-driven searches.
generation data centers, Google will likely use a mix of quantum
computers alongside traditional von Neumann architecture servers.
This would allow the systems to serve diverse requests and use the
best tool for the job for each search.
To get a taste of how
Google's new search works and the mechanics that could drive the
company's next-gen datacenters, read its conference paper, entitled,
2009 Demonstration: Binary Classification using Hardware
Implementation of Quantum Annealing" (PDF).
quote: Right. But how would the system know if the person on the picture is the real one or a twin?