Slashdot Mirror
Quantum Computer Demoed, Plays Sudoku
prostoalex writes "Canadian company D-Wave Systems is getting some technology press buzz after successfully demonstrating their quantum computer (discussed here earlier) that the company plans to rent out. Scientific American has a more technical description of how the quantum computer works, as well as possible areas of application: 'The quantum computer was given three problems to solve: searching for molecular structures that match a target molecule, creating a complicated seating plan, and filling in Sudoku puzzles.' Another attendee provides some videos from the demo." Anyone want to guess how long before "qubit" gets compressed to "quit" (as "bigit" became "bit" in the last century)?
There are fast and almost-exact algorithms for Traveling Salesman problem that are good enough for practical purposes.
I think you've missed the point of the Traveling Salesman problem. It's a theoretical mathematical exercise, not a practical issue in mass transit or shipping. The important bit is that we could solve all sorts of other, more interesting problems if we could solve that one.
Things do get more complicated when multiple qubits are strung together but they still represent zeroes and ones. A three-qubit system can be described by eight complex numbers that keep track of the probability (called "amplitude") of the states 000, 001, 010, 011, 100, 101, 110, and 111. Qubits are bits.
If you want to be annoying and say, "What about a three-state system, where instead of dealing with spin up and spin down, you get spin up, spin zero, and spin down?" then you would have a valid point. But nobody calls such a system a "qubit," any more than we could say that an ordinary electrical circuit holds an ordinary bit if it's allowed to assume three distinct voltages instead of just the usual "high" and "low."
Good job getting one of the first posts, though.
And to avoid the massive worldwide suicide of voice-recognition software who suddenly log-out the computer, in the mid of the dictation of some research paper...
Dear aunt, let's set so double the killer delete select all.
"Sufficiently advanced satire is indistinguishable from reality." - [Tips: 1DrYakQDKCQ6y52z6QbnkxHXAocMZJE61o ]
True, but crosswords always annoy me - half of what you need to know is useless trivia: "Rebel Without a Cause Co-Star"... who cares? Okay, yeah, I know it's Natalie Wood, but that's useless knowledge.
W..w..W - Willy Waterloo washes Warren Wiggins who is washing Waldo Woo.
and lots of smoke is there, too.
Milk and water: take something which is new, but not be interesting scientifically, mix it with something old and dilute it. Shake it for some time. Smoke: Add nice words and senseless technological complications. Claim that your system (although you are doing basic reseacrh which could not be more of from implmenting that) will solve the problems of the world put in your own hand to choose the problem size you want to do this with (If you write for IEEE on Image Processing wou will also not choose a picture where your algorithm fails). Smoke is necessary if you want to use milk and water and cover up that the things you are mixing with are not very new.
Disclaimer: I come from the field. Before I come to my critics, I have to say that I am impressed with DWave having this System developed in this way and I believe that something will come out, probably good research; and maybe eve a working QC.
Regarding the talk of Dr. Geordie Rose:
* he says, they have something but they do not want to compare it to the other approaches, which this time he at least gives an credit, claiming that the others try it differently. This is done to guide the audience away from topic of coherent entangement. I really miss an simple spectrocsopic measurement of their system or some of the things you can do in AQC.
* And, please. As far as I understand the 128 Control lines are used for DC biasing of the coupling SQUIDS. I'd like to see a calculation of the influence of the 1/f noise of the Spectrum of the Hamiltonian for a realistic algorithm.
* He claims that building a complex system out of things you don't understand and enhancing it is more promising than enhancing the single thing and composing it. He says they qould use a quick and dirty approach to it. He misses to mention that they are the only ones seriously doing so. (a fundamental issue about insulator materials used was found out, exactly because one of the leading groups examined a single qubit very careful)
* He brings it into a subtle connection to a technology long existing (RSFQ, which is not used as far as i can see on DWaves chip and Dwave has not much to do with this technology - only that some people working previously on RSFQ now work there) and presents it as something where their research is useful for. This is done deliberatly to stun the audience who most likely have not heard of it. He says that superconductting computers are fast, but indeed the AQC itself is slow. How slow, depends on the algorithm which you use.
* I would classify DWave as a hardware-company. Why all this glittering software around. For the people who want to use an QC it does not matter if you pack it nicely into an SQL server. Call me conservative and square, but somebody showing animations in a technological demonstration frontend has in 50% of the cases something to cover up.
I understand that all this is necessary to impress possible investors. As a scientist I'd be more impressed about a cond-mat preprint where DWave describes the performance of the system in detail. Actually I can't expect it...
The first digital computer systems did not solve anything "amazing" but the fact that they solved anything at all was the amazing bit.
Quantum computing is very new (in the physically exists sense) and the fact that they figured out how to build, program, and extract the solutions for some, albeit relatively simple, problems is a major step forward.
Once the understanding is complete enough and reliable enough then the really tough problems will be sure to follow.
I don't know what you're trying to get across, but you're either wrong or not explaining it well.
Qubits are are not just bits. Qubits are bits in a quantum superposition, and as such do not 'assume' a state from zero to one, but are, instead, all such states at once(1), just like the GP said. (Or all such states in different universes, if that's the interpretation that floats your boat.)
The probability may be writable by physicists, but the actual state of a qubit isn't. (At least not before the calculation is over and it is measured.)
That isn't something that's mildly important, that's how qubits work. Once we actually have multi-qubit quantum computing, and possibly this is going on in this machine, the qubits won't hold 'all' states, but instead specific patterns of states, with different areas being different probabilities. Like interference pattern in a two-slit experiment, where there are likely areas (well-lit), unlikely areas (grey), and impossible areas (unlit). Qubits would 'interfere' with each other until only one (or a few) areas were lit (or dark), and that's the answer. Or, at least, close enough to the answer that a tiny bit of classical checking can nail it down.
There's a quantitative difference between something that holds values that can be measured, and something that holds values that cannot. qubit!=bit
If corporations are people, aren't stockholders guilty of slavery?