In any case, most of these claims have involved quantum computers that do not solve a certain algorithm, but simply behave as a quantum computer. For example, Google's claims are based on what is called 'random quantum circuits', what exactly is how it sounds.
In his own angle is a company called D-Wave, which makes hardware that depends on quantum effects to perform calculations, but is not a general quantum computer. Instead, the collections of Qubits, once configured and initialized, are left to find their way to a land energy state, which corresponds to a solution to a problem. This approach, called Quantum Anlling, is most suitable for solving problems finding optimum solutions for complex planning problems.
D-wave was probably the first company that it “we can perform better than classic” followed by an “Oh no you can't” from algorithm developers, and since then it has usually been much more careful. In the meantime, a number of companies have placed the computers of D-Wave to use problems that match where the hardware is most effective.
But on Thursday, D-Wave will release a paper that again claims, as the title indicates, “Beyond Classical Computation”. And it will do it on a problem that does not contain any random circuits.
You sing, Ising
The new article describes the use of D-Wave's hardware to calculate evolution over time of something that is called an Ising model. A simple version of this model is a two -dimensional grid of objects, each of which can be in two possible situations. The state occupied one of these objects is influenced by the state of its neighbors. It is therefore easy to place an ISing model in an unstable state, after which values of the objects will turn in until it reaches a stable state with a low energy. However, because this is also a quantum system, random noise can sometimes turn bits, so that the system will continue to evolve over time. You can also connect the objects with geometries that are much more complicated than a grid, making complex behavior possible.
