[extropy-chat] D-Wave premiere of 16 qubit processor
scerir at libero.it
Tue Feb 13 22:50:52 UTC 2007
> Anyone there at the demo?
<<Just came back from the D-Wave announcement. Unfortunately I had to leave
at the start of the Q&A session. Here's what I took away from it (minus the
- They claim to solve integer programming problems on their system. Geordie
Rose was careful to explicitly say that they are only hoping to see a
quadratic speedup. Herb Martin (the CEO) wasn't quite as careful in his
opening remarks but then he's the "suit". Geordie said that their current
chip is not a universal QC (presumably because their space of Hamiltonians
is limited) but with some work they expect to be able to make it universal.
- Geordie said compared their approach to the efforts in academia as similar
to Celera and the Human Genome Project. He said they were trying to get
something that would scale fast and worry about about the quality of the
qubits later. He contrasted this to the academic community's efforts to
achieve fine control over qubits before scaling up. They say that they hope
to reach 1024 qubits by the end of 2008.
- They demoed 3 "real-world" problems where they used their system as
essentially a blackbox IP solver.
- Searching for protein matches. Given a protein try to find the closest
match in a protein database. They serially fed all the items from the
database to the chip and asked it to score the match against the given
protein. They said it was solving a maximum independent set problem.
- Finding the best seating arrangement at a wedding reception. Given
constraints on which people can't be seated together and who wants to sit
together, find the optimal arrangement.
- Solving a Sudoku puzzle.
- At one point Geordie quoted you [S.A]. He excerpted a paragraph from your
SIGACT article (the one where you talk about generating Shakespeare's 38th
play) and mentioned your proposal of the inability to solve NP-hard problems
as a physical law. As far as I can remember, the only other computer
scientist he quoted was Gordon Moore so you're in good company. >>
But, wait, this seems more interesting ...
<< I'll pose this puzzler: If quantum computers
are more efficient than classical ones
then why didn't our brains evolve to take
advantage of quantum information processing? >>
More information about the extropy-chat