The impact of quantum computing on data science and security
First of all, to understand what quantum computing is and how a quantum computer will (possibly) impact our lives in the future, we need to compare it with a classical computer, and understand how it processes data.
Classical computer and logic gates
A classical computer operates with bits, that is, binary units of memory that can receive values 0 or 1. All information, including every character and pixel of light that is reaching your eyes right now, is stored on some electronic device (on your computer or in the cloud) in the form of these bits. More specifically, they can be processed through logic gates, that is, physical structures or a system that receives a set of these bits and returns another bit, 1 if the condition is satisfied or 0 if it is not. In the programming area, we call these functions Boolean operators (name given due to the studies of George Boole, who published very important works in algebraic logic).
Briefly, there are 8 types of logic gates that are used to build a universal computer as we know it today. That is, our cell phones, laptops, SmartTVs and pocket calculators use circuits that have these gates in their construction. Each gate has a symbol and returns a different bit based on the input bits. If you are curious to know more, here is some very explanatory material on the subject . Computational algorithms, physically transcribed on our motherboards, memories and processors, use different combinations of these logical gates to process data and information. This is the secret of classical computing.
Ok, but where does quantum computing come in, and what makes it so different for companies to invest billions in its development?
Quantum computing
The term “quantum” was coined by Max Planck at the beginning of the 20th century. XX to describe the energy of radiation that was no longer presented in a continuous form, and now needed to be “quantized”, that is, divided into small pieces or packages so that theory agreed with experiment in some areas of physics. That said, the main difference is precisely in the way a quantum processor operates its data bits.
In fact, a quantum computer operates with what are conventionally called Qubits, and its main (and drastic) difference is that a Qubit can store values of 0, 1 and also a mixture, called superposition, of two states 0 and 1. This third state, which is a direct consequence of quantum physics (part of physics that studies very small systems, on the atomic scale), is what makes it possible to create new logic gates, in addition to the 8 described above, and which algorithms of quantum computing are so powerful.
In other words, classical gates operate with classical bits, which store only 0 and 1, while quantum gates operate on good Qubits, which, in addition to states 0 and 1, have additional properties that are superposition and entanglement.
Superposition
I believe that the most widespread way to explain the phenomenon of superposition is the Schroedinger's Cat thought experiment . Truth be told, Prof. Schroedinger, at the time, thought of this experiment as a counterexample to demonstrate that the current interpretation of quantum mechanics did not make sense. Years have passed and the story of the living and dead cat is now used as a t-shirt on freshman physics courses around the world.
Without going too long, the question is this: Quantum physics is described mathematically through a series of differential equations. These equations, which describe physical states and other properties of matter, have more than one satisfactory solution and describe some experiments well. It turns out that, mathematically speaking, if such a differential equation has a solution A, which describes, for example, the position of an electron; and a solution B, different from A, it is possible to demonstrate that A and B are a solution, at the same time, for that equation. And this is where, in my opinion, all the confusion begins. The current interpretation most accepted today for these equations is that the particle exists in a superposition state, that is, in states A and B at the same time, however it collapses into one or the other when a measurement is made.
Interlacing
Well, as if the strangeness of the concept of superposition wasn't enough, some particles also have a concept subject to it, which is entanglement. When a pair of particles is created, they share some properties in common. In a way, it's as if the same solution A & B were valid for both of them at the same time, even though they are miles away from each other. In some cases, entanglement combined with superposition means that when you measure solution A and collapse the superposition state of the first particle, the second, automatically, (what Einstein called the 'Ghosting effect at a distance ) will collapse to state B without having to take any measurements (remember that, to collapse a state of superposition, it is necessary to take the measurement and in the case of entanglement, the measurement needs to be made on just one of the intertwined particles).
Don't worry if the concepts above seem confusing. They really are. Physicists around the world still debate whether interpretations of quantum mechanics have any more fundamental implications for our reality than currently accepted. However, this does not prevent the results of quantum mechanics from being used in the most diverse applications. And here we come back to quantum computing.
How the quantum computer works
In the same way that classical computers can be built bit by bit, with transistors and the logic gates that I showed previously, a quantum computer is built Qubit by Qubit, with quantum gates that operate these Qubits and allow all the calculations that a common computer would do. (and much more), in theory. In practice, building and maintaining a quantum computer is a laborious task, as Qubits only maintain their quantum properties under very specific conditions, such as very low temperatures, close to absolute zero, and isolated from any source of external disturbance.
IMB coined a term called quantum volume, which somewhat expresses the number of Qubits in a quantum processor versus the error rate. It is a measure of computational capacity. In other words, if the error in quantum measurements persists, even in the thousandths, there is no point in increasing the number of Qubits. Computational capacity would only actually increase when the error was less than 1 millionth, but in any case the two things need to go together.
Challenges of quantum computing
As wonderfully excited as we may be at the news of quantum supremacy achieved by some supercomputer around the world, we must be careful to understand that there are also a lot of myths out there and, more importantly, a lot of development to be carried out in both engineering and technology. systems for these computers. One of them concerns “quantum parallelism”, that is, a quantum computer could perform several operations in parallel and return the best answer to a specific problem.
While a quantum computer can indeed provide you with the best solution to a complex problem exponentially faster than a classical computer, the way in which it does this is intrinsically due to the laws of quantum mechanics, of which we have no understanding. complete yet. Michael Nielsen, author of one of the most cited books in this area, states that if there was a simple explanation for how a quantum computer works, then it could be simulated on a classical computer. But if this could actually be simulated on a conventional computer, then it would not be an accurate model of a quantum computer, since quantum computers, by definition, do not operate in a conventional way.
Therefore, the crucial aspect of quantum computing lies in the way that Qubits are organized into quantum logic gates, which also have a different property from classical gates, which is reversibility. That is, each quantum gate has a kind of mechanism that allows access to previous values, which is not yet possible in conventional computing. We can understand from this that quantum computers do not lose the information that is processed.
Pauli matrices
Almost all quantum physics can be expressed in matrix form. Thus, theoretically, we can describe the Qubits as vectors and the gates in the form of matrices. A gate can operate more than 1 Qubit at the same time. If this is the case, this gate will be represented by a 2×2 matrix. These gates that operate on single Qubits at a time are known as Pauli gates (or matrices), named after the physicist Wolfgang Pauli, who made very important contributions to quantum mechanics.
An easy-to-understand example is the Pauli X gate, which operates in a similar way to the NOT gate in conventional computing.
We say X because Qubits can be oriented in three spatial directions (X, Y, Z), and there is a Pauli matrix to operate in each of these directions. In addition to the Pauli gate, perhaps another of the most important ones that operate on single Qubits is the Hadamard gate. Its main function is to transform a Qubit with a well-defined state (0 or 1) into such a superposition of quantum states.
Quantum computing today
I won't go into detail about the mechanics of each quantum gate, but it's worth saying that, just like in conventional computing, different combinations of quantum logic gates, in theory, can be used to create a universal computer. However, you may have already heard about quantum computers in operation . These machines, despite claiming to solve problems hundreds of times faster than a classic supercomputer, are still not universal equipment, and their chips are created to solve a single task. As of the date of writing this article, there is no news of any company that has created a universal quantum computing chip, but there have been advances and, currently, companies like IBM have in their development laboratories chips with more than 100 Quibits in operation, but for the purpose of solving specific tasks and calculations.
Another thing worth highlighting is that these quantum computers in operation are gigantic machines, which occupy rooms, and perhaps entire floors, to remain in operation.
Decoherence
The phenomenon that haunts quantum processors around the world is decoherence. In the best single-purpose quantum computers, it happens after a few seconds of operation, and is mainly due to the difficulty of keeping the Qubits in their initial states and in superposition. In fact, decoherence is the main reason why we do not observe quantum phenomena in our macroscopic world. If we go up by just a few orders of magnitude in terms of number of particles, temperature, or size, decoherence comes into play and the entire system returns to behaving in accordance with good old classical physics. However, if we remember that ENIAC, the first universal computer , also occupied an entire room, and we consider that Moore's law will also apply to quantum processors, we can be promising in admitting that in a few decades we will have real quantum computing capacity for applications in the most several areas.
Quantum computing and applications
Even with just a few dozen Qubits, universal quantum processors would already be capable of surpassing current computational capacity in some specific problems, such as molecular modeling, essential for the manufacture of new drugs and the study of diseases, route and process optimization problems and categorization of prime numbers. This last aspect is perhaps what leaves many information security managers with nightmares at night.
It turns out that most data encryption and decryption algorithms today work with keys based on prime numbers. Therefore, a universal quantum computer could, in theory, obtain a decryption key in a matter of seconds. The best firewalls and encryption systems would be almost transparent to a universal quantum computer with a few hundred Qubits. The main reason for this is the way quantum processors process data.
Data security
Consider an example of a database with 1 billion rows, where each row contains a name. To search this database, a conventional computer (disregarding optimization algorithms) checks that list name by name, until it finds the desired result. Very simply put, in a quantum computer this list could be stored in a superposition state using Qubits, then the “query” would be applied in order to collapse this list to the desired name in the search, with a single operation. With this simple example we can already see the striking difference in the performance of quantum computers compared to conventional ones, and much more can still be done and is under development as you read this article.
However, there are already ways for managers to defend themselves when quantum supremacy is actually achieved. Since the second half of the century. XX, there are quantum cryptography algorithms that use other (weird) properties of quantum physics to their advantage. An example would be the (famous) case of Alice and Bob. If Alice sends a quantum-encrypted message to Bob, Bob can know if the message was intercepted, as a third-party reading of the message sent by Alice would alter the Qubits sent significantly. This occurs because superposition states, as she said, are altered or collapsed when a measurement is made. I believe it goes without saying that the real scenario is infinitely more complex than the one I presented, but this example was just to highlight that not all is lost regarding information security when (or if) the first universal quantum computers are available.
Quantum computing – Final considerations
Finally, I must highlight that, although it still seems like a distant future, quantum computing is already on the market and is expected to generate billions of dollars in the coming years. I believe that we will still see the paradigm of quantum computing being broken and all the implications that this will bring.
Even though we still don't 100% understand all aspects of quantum physics, we have still been using it for over a century in various electronic devices. As I mentioned in the previous article , the miniaturization of transistors was only possible thanks to quantum mechanical calculations. Therefore, not fully understanding what goes on inside the black box, which are quantum logic gates, will not prevent the industry from investing in and using this technology when it is ready.
There are still many challenges, whether in the development of new materials, hardware construction, programming algorithms, systems development, etc. However, it is good to be prepared for changes, especially in a world that is more and more dependent on data processing, as the world will certainly change when this processing grows exponentially with the advent of quantum supremacy.
Also read: Tactile Internet
References:
https://towardsdatascience.com/demystifying-quantum-gates-one-qubit-at-a-time-54404ed80640
https://towardsdatascience.com/the-need-promise-and-reality-of-quantum-computing-4264ce15c6c0
https://www.ibm.com/quantum-computing/
https://ai.googleblog.com/2019/10/quantum-supremacy-using-programmable.html
https://www.quantamagazine.org/the-era-of-quantum-computing-is-here-outlook-cloudy-20180124/
https://www.techtarget.com/searchsecurity/definition/quantum-cryptography#:~:text=Quantum%20cryptography%20is%20a%20method,secret%20key%20can%20decrypt%20it.
https://tecnoblog.net/especiais/joao-brunelli-moreno/eniac-primeiro-computador-do-mundo-completa-65-anos/
Who is Aquarela Analytics?
Aquarela Analytics is the winner of the CNI Innovation Award and a national reference in the application of Artificial Intelligence in industry and large companies. Through the Vortx platform and the DCM methodology ( Free e-book download ), it serves important clients, such as: Embraer (aerospace), Auren (energy), Scania and Randon Group (automotive), SolarBR Coca-Cola (food), Hospital das Clínicas (health), NTS-Brasil (oil and gas), Telefônica Vivo (Telecommunications), among others.
Stay tuned for new daily publications from Aquarela Analytics on Linkedin and by subscribing to our monthly Newsletter
Comments
Post a Comment