Quantum Computing, in a Nutshell

Well, let me give it to you straight. Most explanations of Quantum Computing use Bloch Sphere diagrams that will bore you to death as they are too technical or make you think there must be a Quantum Computer available now that will allow us to crack every password in existence in two years! Neither of those is true. I promise to provide you with a 'real' explanation of Quantum Computing!

Classical Computers (like the one you have in your laptop or smartphone) store information in Binary Format or as 'Bits'. Each bit can be either a '0' or '1' at any given moment but not at the same time. Everything that your computer does: running games, compressing files, rendering videos, etc., is based on a massive number of these Binary Decisions happening in very very short time frames. When you think of how much can be created on such a small base, it seems a bit ridiculous - however, there is a limit!!

A Qubit is not simply '0 and 1 at the same time'; this explanation makes it seem pretty simple, but actually, it is much weirder; therefore, it will require more explanation.

Qubits are the basic unit of information in a quantum computer. When people try to explain to you how a qubit works, they will often say that a qubit can be "0 and 1 at the same time," which is somewhat misleading. A better way to think of a qubit is as being in a superposition, which is a weighted combination of 0 and 1. Think of it like a coin that's spinning in the air. It's not heads, it's not tails, but it has some probability of landing on either. When you measure it (i.e., when you look at it), it collapses to a definite 0 or 1. The quantum magic happens before the measurement.

Now, when you have 300 qubits all connected together, that allows your quantum computer to represent 2 300 (about 10 90 ) different states simultaneously. That's more atoms than exists in the observable universe. A classical computer would need to check those states one at a time. A quantum computer can, in some sense, explore all of them at once. This is called quantum parallelism. Sounds like a superpower, right?

That's true to an extent but has a very notable limitation. When you measure a quantum system, you only get one answer. One classical bit of output per qubit. All that super position collapses down to a single result. So, the art of developing quantum algorithms is all about how to structure the calculations in such a way as to generate erroneous answers that cancel each other out (over a process called quantum interference) and generating results that can be amplified. It's like tuning a radio - you suppress noise, boost signal. This is legitimately hard to do, and it only works for specific types of problems.

The other big quantum ingredient is entanglement. The states of two qubits can be entangled, meaning that if you measure either one, you instantly constrain the state of the other qubit no matter the distance separating the two qubits. Einstein famously detested entanglement and referred to it as "spooky action at a distance," yet there are countless experiments which demonstrate that this is a real phenomenon. Entanglement lets quantum computers correlate information across qubits in ways classical systems simply can't replicate efficiently. It's what gives quantum algorithms like Shor's and Grover's their edge.

Shor’s algorithm is able to exponentially factor numbers that are much larger than anything we can do with classical methods. Factoring is a common assumption used in modern encryption, making this a big deal. Grover’s algorithm, on the other hand, provides quadratic speedup for searching unsorted databases. If a classical method takes N to finish a search, Grover can do it in √N; so not as impressive as Shor, but it's provably optimal for that class of problem, which is cool in its own right.

Should we be concerned about quantum computing today? No, not right now. As of now, researchers refer to today’s quantum computers as Noisy Intermediate Scale Quantum(NISQ) devices. The ‘noisy’ part of the name has to do with qubits being extremely sensitive to outside interferences. Things like thermal vibrations, stray electromagnetic fields or even cosmic rays will affect the quantum state of the qubit, causing it to lose coherence. In order tokeep qubits stable, IBM and Google cool their chips to near absolute zero; this is approx. 15millikelvins. Deep space is much warmer than this temperature. The engineering hurdles associated with this work are significant.

To use Shor's algorithm to break modern RSA encryption, we will need thousands of logical qubits. A logical qubit is a qubit that has undergone error correction and is a combined product of many physical qubits. Today's best machines have hundreds of physical qubits, and error rates are still too high to string them into reliable logical ones at scale. Therefore, we are not yet there. However, the trend is upward - every few years, there is an increase in the number of qubits and a decrease in the error rates. It isn't Moore's Law but there is a similarity.

The more feasible applications of quantum computation in near term are in quantum simulation. Simulating molecular behavior, chemical reactions, and the properties of materials is a realm where the underlying physics is quantum physics and classical computers have to approximate brutally. A quantum computer simulates quantumphysics naturally. Using quantum-computational methods of simulating nature with the associated speedups would greatly enhance drug discovery, battery design, material science, and many others that have direct implications for the real world.

My mental model for quantum computing is similar to this: a quantum computer cannot be conceived of as a faster classical computer but instead as a different form of computation for a group of problems where the effects of quantum mechanics provide an indisputable advantage. For most of what computers do - serving web pages, running spreadsheets, training neural networks - classical hardware wins, and will keep winning. But for the narrowcorridors where quantum shines, the speedups aren't incremental. They're exponential; this by itself, should capture your interest.

We are at a very early stage of development: The Hardware is quite raw, system programming models are still too complex, and almost all "quantum advantage" claims should be approached with caution; However, the theoretical basis appears sound, there is true engineering progress being made, and the long list of significant problems that Quantum computing will eventually be capable of resolving is undeniable; All of this makes for a very unique and important reason to continue to follow the development of Quantum computing closely.