QuBits: Ten Things You Should Know About Quantum Computing
Big tech companies like Google and IBM are pouring billions of dollars into quantum mechanics research with the hope that these investments will keep themselves in the international race to build the first universal quantum computer.
Compared to today’s classical computer systems, quantum computers will have the ability to encode much more information and solve long, complex problems in minutes, rather than the years, and possibly decades, it would take a classical computer system.
So how exactly do these quantum computer systems work? We reached out to Michelle Lollie in LSU’s Quantum Science and Technologies Group to explain 10 things we should all know about quantum computing.
Lollie is a second-year graduate student and member of LSU’s Quantum Science and Technologies Group in the Department of Physics & Astronomy. The group conducts research on atomic, molecular, and optical physics, the foundations of quantum mechanics, photonic bandgap and metamaterials, quantum information theory, quantum complexity theory, quantum error correction, quantum optics, optical quantum computing, quantum sensors, quantum imaging, and relativistic quantum information theory.
First things first, there is no such thing as a universal quantum computer, yet. Theoretical physicist Richard Feyman is often linked with the phrase universal quantum computer, but he was actually referring to a universal quantum simulator in his seminal 1982 paper on the subject. But, I’ll let my colleagues hash that one out. Moving along, there are several universal quantum computer hopefuls in the game (think Google, IBM Q, Intel, Rigetti) that use quantum bits on superconductors to process information for specific problems but we have yet to be able to do “all the things” because these systems use a relatively small amount of quantum bits. More on that later.
Quantum computers use the concepts of superposition and entanglement. Think of classical bits versus quantum bits or qubits. Consider the classical bit states of zero or one. Now consider a qubit state. It can be a zero, one, or every fluid state in between simultaneously, until observation. Once a qubit state is observed, its environment is disturbed and it collapses into, or basically goes back to “being,” a zero or a one. This is the concept of superposition. Now take the one qubit and connect it to another one. If they are connected in a particular way, such that their intrinsic properties are shared, they are said to be entangled. If you observe something about one, you automatically know this information about the other. This infers that in describing one qubit, information about the other is known, even if they are then displaced light-years apart. This intrinsic information can even be teleported between the two qubits. And no, the information is not teleported faster than the speed of light because, in order to extract the realized teleported information, classical bits are required, thus slowing down the process. Einstein is sighing with relief somewhere.
Computational power for quantum computers goes as 2^n, where n is the number of qubits. It can be roughly shown via a linear graph that the amount of qubits on superconducting chips doubles about every six months as a function of time (2016-2018). So the processing power increases yearly as 2^(2^n). Let’s say you have three qubits, this implies a processing power eight times that of a classical computer with three bits (2^3). I’ll let you do the math for 60 qubits. Imagine one million qubits!
So what is all the hoopla about entanglement? Well, entanglement is shy. It cannot be observed directly. Quantum systems that exhibit entanglement must be kept in pristine environments in order to harness this processing power. If even one photon, a light wave-particle, interacts with the system, the system will collapse into discrete zeros and ones no longer exploiting the necessary superposition or any entanglement. Furthermore, entanglement would have to be maintained for practical time periods like hours, days, etc. Imagine trying to put yourself in a vacuum (just go with it despite self-preservation interests). How would you go about it? How can you remove all of the air and light around you and for how long could you do it? Well, this is the case from the qubit perspective. This is not easy to do in a lab setting, let alone the real world.
Ideally, a universal quantum computer will be virtually hack-proof. If an eavesdropper were to secretly try and access a secret code, the resulting message would be distorted and the recipient would know. However, we live in the real world where nothing is certain (kudos to Heisenberg for quantifying this), so there are ways to access secure info even with a quantum computer. This is interesting-- if a distorted message is received, was it hacked or just acted on by its environment? Researchers are actively working to reduce the uncertainty due to environmental factors and increase the fidelity of secure messaging.
This leads to the characterization of uncertainty and accounting, which is error-correction. Mathematical probability plays a fundamental role in quantum mechanics, so it’s not a secret that there is an area in quantum computing known as error-correction. For every parcel, if you will, of information stored in qubits, there has to be an error-correcting code to account for the environmental and computational loss of this information as it is stored as well as the method of its communication and realization. Remember the game telephone? One has to think of a message, store it in memory, and share this information with another individual securely ensuring the accuracy of the message is communicated. Not so hard for two people, right? Well, what about ten people? 100? 1,000,000?
Yes, quantum computers are able to hack public key encryption. Public key encryption is used ubiquitously to secure high-level information (think bank cards, government information, etc.). There are even quantum hackers (yikes!). If all classical computers existing today (including your laptop) were combined yielding very high computational processing power, they would be unable to crack public key encrypted information if the key is long enough, but one universal quantum computer could potentially crack this information in a matter of milliseconds! Fret not, we don’t have fully universal quantum computers yet. These devices will be able to run Shor’s algorithm, a set of directions to factor the large numbers that we use to make the public keys that secure our data. Enter quantum hackers - they are on our side helping us to understand the strengths and weaknesses of our quantum systems Also fret not because…
We need millions of qubits to build universal quantum computers. For reference, IBM Q and Google boast quantum computers with an order of ~60 qubits. Although they are doing interesting work in building their quantum devices and utilizing them for small applications, scalability is the saving grace here when it comes to information security. In order to run Shor’s algorithm, the quantum computer has to be large indeed (size = number of qubits)
There are a few popular materials or approaches used to build the hardware that will power universal quantum computers. The most well known (some realized) are superconductors, semiconductors, all optical devices, and ion traps. If you’re interested in a quantum engineering career, look into atomic and optical physics!
Beam Me Up Scotty - Quantum Teleportation is not Star Trek. But it is useful for moving around quantum data. On the flip side of the quantum computing coin is advanced communication. Enter the quantum internet! Information is teleported between networks and computational devices (classical and eventually quantum) across the quantum internet. The Chinese satellite Micius is an integral part of this endeavor. The satellite is the first quantum satellite ever. It has used entanglement between photons to distribute secure quantum keys enabling a secure call between ground stations in Vienna and Beijing, a distance of about 4600 miles!
It is exciting to think about the processing power and security of quantum computers, but we are still years away from having these systems in our homes and offices. Some researchers predict that we will be able to buy our own quantum computers by 2050. For now, we will have to be content with observing big tech companies and top universities, like our own, as they stake their claim in the global race to conquer the quantum domain.