@TheJoe: Microsoft beat Google to it. Do a search for Microsoft Station Q research
Though I don't know if Microsoft has an actual computer they are calling a quantum computer.
Do you Microsoft?
I don't really understand the details of quantum computing that well, but I know a quantum computer is not simply a faster conventional computer. But it can do things that conventional computers are not very good at. For instance, many machine learning algorithms are dependent on an operation called gradient decent to improve the accuracy of the prediction model.
The problem with gradient decent is the decent accounts only for the current position on the gradient. Because of this, it is very possible to converge to a local minimum and have a less then optimal model as a result. Quantum computing somehow overcomes that.
This is probably why Google researchers are trying to apply quantum computers to the problem of machine learning. Many researchers under the employ of Google are involved in trying to push the limits of machine learning, because so much of Google's core business depends on it. So they are trying many different ideas, some may be more realistic then others.
A company near me sold a quantum computer to Lockheed Martin.
It's hard to determine what is meant by speed compared to a conventional computer. Is it compared to a single transistor? Quantum state means it has all possible values, so how do you program a quantum computer ?
-- - -
Energy consumption is an important factor in super computer usage. Its an additional cost that must be considered for each program run.
There are two main obstacles that need to be overcome for future improvements :
Can the quantum computer run Crysis? I think that is an excellent question. What about it channel 9 people? Can you walk over to Microsoft Research station Q and ask them what is up with Quantum computing? I would appreciate a fresh and clear take on where the world is in developing functioning quantum computers. Thanks!
"While D-Wave's hardware is better at dealing with structured code, it runs neck-and-neck with the "fake" system when tackling random problems. "
What I don't get is it seems they compared the quantum computer to a conventional computer running probabilistic gradient algorithm. This means it has a chance to get the optimal result, but it isn't guaranteed. Meaning if you are looking for something that produces the optimal result, there is a chance that the algorithm produces the wrong result. This probabilistic approach is a common technique to approximate solutions to problems that would be NP otherwise.
But, what I don't know is if the quantum algorithm has the same limitation (ie. can it produce a wrong answer, sometimes)?
Comments have been closed since this content was published more than 30 days ago, but if you'd like to continue the conversation, please create a new thread in our Forums, or Contact Us and let us know.