July 3, 2013
D-Wave Takes Quantum Computing A Step Forward With Quantum Optimizer
redOrbit Staff & Wire Reports - Your Universe Online
Quantum computing company D-Wave Systems has made impressive inroads in demonstrating how their quantum optimizer can have real-world applications. The company has sold their system to Google and Lockheed Martin in recent months, and appears to have made another step forward in demonstrating that their machine is not performing classical simulated annealing.
A set of magnets provides an example of annealing. Magnets can arrange themselves to either point up or down. At high temperatures far in excess of the ground state, each magnet arranges itself randomly, either pointing up or down. Because the energy difference between the two states is small, the magnets can move up and down in response to small changes in the local magnetic field caused by neighboring magnets reversing. However, as the magnets are cooled, the rate of flipping slows, and the influence of the direction of the neighboring magnets becomes more significant. In this case, where the only magnetic field is due to the magnets themselves, they begin to pair up to minimize their total joint energy and lower the total magnetic field toward zero. In other words, each magnet tries to minimize its energy with respect to its surroundings.
This process, known as annealing, can also be simulated using a computer, and provides a good way to solve highly complex problems. However, simulated annealing is theoretically no faster than any other classical mechanism for calculating solutions to problems - although it may be more convenient.
This is basically what D-Wave's system does by using superconducting loops to generate tiny individual magnets where each ones orientation depends on the direction in which the current circulates. The magnets are then coupled to each other so their orientations influence each other. To solve a problem with this group of magnets, the problem must be restated so its solution is the magnets' lowest energy state.
To get to that point, the magnets' orientations are initialized in a well-known way, and the system is placed in the ground state for that configuration. Slowly, the environment around the magnets and coupling among them is modified so it resembles the problem. If done correctly, the magnets may change their orientations, but never leave the ground state. By reading out the magnets' final positions, a solution to the stated problem is obtained.
If this happens to be a classical process, as described above, then the approach is no faster than a classical computer. But if the magnetics are behaving in a quantum manner, it might be a quantum computer and could offer a faster way to solve problems.
The researchers analyzed how the coupling between the magnets created a ground state. The configuration used in the study consisted of four inner magnets arranged in a diamond, so that each magnet was coupled directly to two others. Each of those was coupled to one additional magnet, which were not coupled to each other. This configuration appears to be set up such that the four inner magnets always have the same orientation, while the outer magnets are free to arrange themselves as they see fit, ArsTechnica's Chris Lee explained.
The results included a strange set of 17 possible ground states, most of which can be reached in steps of single flips of magnets, with the exception of the last ground state, which required all four inner magnets flip at the same time.
In a classical simulation, the magnets can sample many different states. However, if it happens to flip into this last ground state it becomes trapped there. Moreover, once there, the outer magnets become trapped in a single state as well because all other configurations have higher energy. Once in this isolated state, the magnets could get out by flipping all four inner magnets, but the isolation and lack of environmental variance (since the outer magnets can't flip either) means it is less likely to flip out of the state than into it.
In the quantum description of these events, this doesn't happen.
After setting up the ground state, the researchers began trying to move to the solution state by varying the environment. As soon as they did, the ground state split up, and the isolated state where things get stuck raised the energy above ground state. Since everything was kept in the ground state, the probability of entering the isolated state reduces sharply.
But this is different from the classical case, in which there was no way to break up the ground state, meaning the D-Wave system is not a clearly classical device.
"Here we present an experimental signature which is consistent with quantum annealing, and at the same time inconsistent with classical thermalization," the researchers wrote.
What this means in practice will likely require additional work, but the company appears to have made another step forward in demonstrating its system can deliver quantum computing solutions.