Primitive quantum computers may already outperform standard machines for very specific tasks
Quantum bits (qubits) are the building blocks of quantum computers, but putting enough of them together in the one device to run computations like those expected in a standard computer is difficult to say the least. But now researchers have come up with a way to use even primitive quantum computers to run calculations that can already outperform the capabilities of classical computing for very specific tasks.
There already exist a number of quantum processors used for research, but these have a very limited number of qubits at their disposal, partly because the fragile nature of quantum entanglement that encodes the qubits makes it extremely difficult to maintain many more. For example, IBM's cloud-enabled quantum processor boasts just five qubits and, as a result,can only run limited algorithms and experiments. This means such devices aren't very useful if you want to perform tasks that your average desktop computer can already do very well.
However, for solving extremely specialized problems, quantum mechanics is leaps and bounds ahead. In this vein, researchers from the University of Bristol and the University of Western Australia are studying the processing capabilities of smaller, simpler designs as precursors for large-scale quantum computers, but also as processing units in their own right.
"A quantum computer is a machine designed to use quantum mechanics to solve problems more efficiently than any possible classical computer," said Dr Ashley Montanaro from the University of Bristol's School of Mathematics. "We know some algorithms that can run on such machines and it's an open and exciting challenge to find more. But most of the quantum algorithms we know need to be run on a large-scale quantum computer to see a speed up."
But examining smaller, precursors to a large-scale quantum computer could help accelerate development of more power machines and even be used to perform valuable work on their own. The questions facing researchers face in this regard involve discovering what these simple computers can usefully do and what are the minimum parameters of performance and construction they require to achieve those tasks? To address these questions, the scientists looked at ways to use a basic quantum processor to illustrate a phenomenon known as the quantum walk.
In essence, the quantum walk is a quantum mechanical analog of the classical random (or "drunken sailor's") walk, where the current state of a walker staggering along is described by a probability distribution (a probability assigned to each possible direction of movement in a given space) over positions. That is, how likely it is, given the number of possibilities in a finite set, that the walker will move in a particular direction.
The overriding difference in the quantum version of the random walk, however, is that it is applied to the random movement of a particle which may exist at any and all possible positions simultaneously, and in various dimensions. This quantum capability has thus allowed other researchers to demonstrate alternative ways to think about how differently full-scale quantum computers may operate, the information density levels they may contain, and how best to create useful quantum algorithms to exploit these facts.
Using a quantum circuit built on a two-qubit photonics quantum processor, it would seem that the team has discovered that it may be possible to harness the power of quantum processing far sooner than first expected.
"An exciting outcome of our work is that we may have found a new example of quantum walk physics that we can observe with a primitive quantum computer, that otherwise a classical computer could not see," said Dr Jonathan Matthews, from the University of Bristol. "These otherwise hidden properties have practical use, perhaps in helping to design more sophisticated quantum computers."
Such algorithms are very useful for modeling such things as Brownian motion – the seemingly random movement of particles suspended in a fluid – and Boson Sampling – an experimental system that may help science move towards photonic elements of quantum computing.
"It's like the particle can explore space in parallel," said Xiaogang Qiang, PhD student in the School of Physics at the University of Western Australia (UWA) who implemented the experiment. "This parallelism is key to quantum algorithms, based on quantum walks that search huge databases more efficiently than we can currently."
The upshot of this research is that a new link has been established between quantum walks and computational complexity theory that shows specific tasks could ultimately demonstrate quantum supremacy over classical computers.
The results of this research were recently published in the journal Nature Communications
Source: University of Bristol