Quantum supremacy: what’s all the fuss about?
By Tarek Schoehl
On October 23rd, 2019, Google finally released its paper claiming to have reached ‘quantum supremacy’, four weeks after rumours and leaks pre-emptively announced the achievement. Google’s quantum computer, utilising their brand-new ‘Sycamore’ chip, can complete a problem in 200 seconds, a feat that would supposedly take supercomputers 10,000 years!
But what does ‘quantum supremacy’ mean?
Quantum supremacy is defined by the ability of a quantum computer to carry out a specific problem impossible for a classical computer, due to the insane lengths of time it would take to calculate the result (think along the lines of 10,000+ years). Theoretically, a classical computer could solve any problem given enough time, but that’s just impractical for obvious reasons.
Oh, and keep in mind when I say “classical computer”, I really mean supercomputers, which can be as large as 2 basketball courts, a fair bit faster than the phenomenal Billy B computers. So how exactly is this quantum computer supposed to beat a normal computer?
While both computer types use bits (a string of zeroes and ones) to carry out their computations, classical computer bits can only either be a zero or a one. This is where the quantum computer has a massive advantage: it computes using quantum bits (qubits), which can be either a zero, a one or both a zero and a one at the same time.
The only thing that matters here is that the more possible states (states being a one, a zero and a zero and one at the same time), the more computing power you have. Increasing the number of qubits increases the possible states much faster than standard bits do. This is what makes a quantum computer so powerful.
What gives this so called quantum computer the right to be called one, and what does one look like, or even consist of? The definition of quantum is “the minimum amount of a physical quantity”, whatever this physical quantity may be. Some quantum computers consist of an array of atoms, on which computations are carried out on. Let’s try to wrap our heads around this: we are talking about manipulating singular atoms in order to solve a complex problem. Just for measure, atoms are around one million times thinner than the human hair!
Quantum computers manipulate singular atoms to solve complex problems
However, the quantum computer Google used is slightly different, as it uses superconducting circuits, which is closer to what a normal computer is made from. Without subjecting you to a lengthy discussion of what ‘superconduction’ consists of, at its most basic it requires things to be extremely cold. Cooling these chips down sufficiently makes the electrons in said chip turn (condense) into a quantum state.
These quantum states are incredibly sensitive to outside interference, with even a slight vibration whilst the computer is running resulting in a worthless calculation.
This intense sensitivity, combined with the high-tech equipment required before even considering building a quantum computer, all contribute to the difficulty and importance of Google making this claim of achieving quantum supremacy.
IBM, one of Google’s rivals, has said that Google’s claim of beating a classical supercomputer by such a margin and hence having established quantum supremacy is ill-founded for two reasons:
A) IBM believes that if the most powerful supercomputer were to be fine-tuned for the problem Google sought to solve, it would actually be able to solve the problem in a mere two-anda-half days; quite the improvement from the 10,000 that Google claimed it would take.
B) The problem that Google sought to solve is very much an isolated one and solving it is not going to help solve other problems in the same way.
To understand what is meant by B) I should talk about classes of problems: different problems are categorised in ‘sets of difficulty’, where if you can solve one problem in the set much quicker, then you will be able to solve all of the problems of this set quicker as well. Whilst Google may have been able to solve this particular problem much faster than a supercomputer, it cannot, as of yet, use its quantum chip to solve other problems in the same short amount of time. This makes Google’s quantum chip very useful for this one problem, but useless in solving very difficult problems, such as factoring integers.
Yes, you heard right, factorising integers into prime numbers is apparently a very difficult problem. This is because we are talking about incredibly massive numbers here, so if you keep increasing the digits, the computer will soon enough reach its limit too.
A quantum computer, on the other hand, uses quantum algorithms to solve these problems much faster than classical computers.
In order to refute IBM’s claim, and retain the quantum supremacy crown, Google needs to increase the number of qubits they can fit into one chip (currently at 54 qubits), increasing the power of their quantum computer and allowing them to apply its superior strength to more complex problems.
Image by Steve Jurvetson, via Flickr
