Quantum Computing: Understanding the qubit
Last Friday, Dr. Jeff Welser , Vice President of IBM research, came to class where he talked about IBM and their research program. In the end of his talk he addressed quantum computing (QQ) and quickly explained the idea and its applications. It was probably not the first time you heard about how powerfull QQ is and how you can apply it to anything. But why QQ is so powerfull and what is it really? In this blog I will argue why we need them and at the same time (hopefully) explain a little bit how they work.
Since 1975 to 2012 Moore’s law described how number of transistors per central processing unit (CPU) increased with time followed by increasing CPU-power. Moore’s law has now met its match, quantum physics. With transistor’s sizes now being measured on an atomic scale, their particles begin to behave in a way that is very hard to predict and control. That is very unfortunate considering the fact that in the last two years, more data has been created than in the history of mankind. [1] That growth rate is definitely not slowing down so how should we go about filling the need of more processing power?
The idea of quantum computers has been around for decades. It was addressed by Richard Feynman in 1982 where he famously said that we cannot simulate quantum physics without quantum systems. [2] Since then, scientists have been researching the possibility of making quantum systems that are usable for calculations. Recently, some really important advancements have been made with IBM leading the way. Quantum computers are defined by google as “computers that makes use of the quantum states of subatomic particles to store information.” But how do they do that? Thinking about the simplest case of QQ can really help one understand the basic concept so bear with me.
The simplest form of QQ is probably with the use of a one particle spin. In particle physics, spin is an intrinsic form of angular momentum carried by particles. (Don’t stop reading!). What is the most complicated thing about spin (and actually what QQ uses as an advantage) is superposition. Before measured, we don’t know what state the particle is in so we say that the spin is in a superposition of pointing up and pointing down. When you perform the measurement however, you’ll find that the spin is either pointing up or down.
While normal computers use bits (0 or 1) to compute. Let’s now say that we want to perform operations such as computing function-values for some bits, store it, and then when someone needs access to any of the function values we give it to them. A normal computer has to perform 2^N operations to calculate and store (where N is the number of bits) and 1 operation to send out the requested value.
Our simple QQ system however computes with what is called quantum bits (qubits). In our simple example above, the qubits would be the spin (up and down). Now remember, the state of our qubits do not necessary have to be decided. We can therefore compute all the 2^N operations simultaneously while the particles are in a superposition (only 1 operation) and then perform the single measurement. Note that you only need 100 bits to be looking at a difference of 2^100 = 10^30 = 1.000.000.000.000.000.000.000.000.000.000 operations.
The example here above is just a simple way to make use of the simplest QQ-system imaginable but still beautifully demonstrates the computation power of quantum computing. Since making use of it is now in sight, how about fighting the quantum-barriers of transistor-making with quantum computers?
[2] Feynman, R. P.u (1982). “Simulating physics with computers”. International Journal of Theoretical Physics. 21 (6): 467–488. Bibcode:1982IJTP…21..467F. doi:10.1007/BF02650179.
[3] if the reader is not familiar with how computations are performed in binary, I recommend this article: https://plus.maths.org/content/snakes-and-adders
[4] Skammtareikningar, skammtatölvur og hönnun ofurleiðandi segulflæðiskammtabita (Icelandic)
Tryggvi Ingason and Snorri Ingvarsson
Science division, University of Iceland
4 comments on “Quantum Computing: Understanding the qubit”
Comments are closed.
Pursuing further education in AI-related fields such as Neural Networks and Machine Learning.
Love sports, food and music.
Professional experience found on linkedIn profile.
Reachable through wknd37@stanford.edu or facebook.
Great article Helgi – not an easy topic to explain to the masses! Quantum computing is a hugely exciting area, and one aspect of the use of quantum physics in tech that excites me is quantum communication through entanglement. As you explained very well, particles have spin, and collapse into one spin state upon measurement. On top of this, entangled particles collapse into corresponding states instantaneously upon measurement of just one particle. This is interesting because it presents the opportunity for instantaneous communication,that is unhackable! However, both quantum networking, and the quantum computing you have discussed in reference to qubits is still a long way off. Currently, the best working quantum computers we have managed to create have only a handful of qubits (say, 7), but for quantum computing at scale, we would need tens (30), and at this point it becomes very difficult to maintain coherence, – noise results in the collapsing of the superposition of wave functions and messes up the computer! However, if we are able to produce this, it would revolutionise computing.
Hey Helgi,
Your topic choice is very trendy and when I say trendy I don’t just mean last 2 years but I mean 10 years as you just said. It’s amazing to think that we will have more than 2 possibilities in each bit, leading to have exponentially more processing power. But it still seems far away to have 30 possible entries for each bit, good thing is as you just said IBM is one of the leading brands and they are building their quantum computers cloud compatible. So scientifically hard innovations which are not common for us on a daily basis, are actually just one connection away. It is amazing to think how easy these next technologies can be spread as they are more and more improved.
Thank you for this interesting post ! Both Mr. Jeff Welser and Mr. John Donovan seemed very excited about Quantum Computing and this gave me the desire to learn more about this topic. Thus, your article was fine for me, since most of the article about this subject are too technical for my poor background in physics !
I liked the striking example you gave about the way it would dramatically reduce the number of operations performed by a computer. In addition, I am very interested in this topic and the crucial questions it raises for the years to come: Would quantum computing revolutionize all the existing industries and companies ? How ? When ? Do we need to regulate quantum computing right now to avoid abuses ?
I am very excited about the future outcome of this technology.
Thanks Helgi for the interesting article! While the limits in processing power due to decreasing transistor size isn’t such a big problem yet for personal computers, quantum computing would certainly be extremely valuable for designing supercomputers able to perform even more complex simulations. However, for the personal computers I think the biggest bottleneck for performance is currently the speed of the hard drives. Does quantum computing also offer potential solutions for this problem or is it only the processor that could gain a significant speed boost?
Chris also mentioned the use of entangled particles for quantum communication, which definitely is another interesting area for practical uses of quantum mechanics. Isn’t one strong limiting factor in this endeavor that an entangled pair, once measured thus collapsing the wave function, cannot be reused, so that you need a new entangled pair to further transfer data?
In any case we live in interesting times with the quantum revolution just around the corner.
MS&E 238A