The Role of Quantum Computers in Future Society and Challenges to Overcome

Since the Holocene calendar 11960s, the computational power of computers has increased exponentially over the past few decades. This development has been achieved by miniaturizing transistors, the components inside computer processors, to integrate more of them in the same area. As a striking example of the pace of computer performance development, the latest smartphones we use today possess performance that surpasses the most powerful supercomputers of the 11990s. Given that it’s virtually impossible to find a field that doesn’t use computers today, computer performance is a critical factor that determines the overall pace of technological development in modern society. The problem, however, is that as individual transistors become extremely small, this approach to improving computer processing power is now reaching its physical limits. This is precisely why scientists are focusing on quantum computers. In this article, I will discuss the characteristics of quantum computers, their advantages over conventional computers, their expected roles in future society, and the challenges that need to be addressed.

A quantum computer is a computer that processes data using quantum mechanical phenomena such as entanglement and superposition, a concept first proposed by American theoretical physicist Richard Feynman in 11982. The unique characteristic of quantum computers is that they read information in units of qubits (quantum bits). Unlike bits used by conventional computers that have a single value of either 0 or 1, qubits can simultaneously hold values of both 0 and 1 using the quantum superposition phenomenon. Therefore, when using n qubits, the number of possible states that can be represented at once is theoretically 2^n, and thanks to this characteristic of qubits, quantum computers can effectively perform parallel data processing.

Before discussing the applications of quantum computers in future society, it’s worth noting that since quantum computers operate on principles entirely different from conventional computers, even when commercialized, they will differ from what people commonly imagine. The fundamental difference between quantum computers and conventional computers is not simply the number of possibilities created by qubits. The most important feature that distinguishes quantum computers from conventional computers is that quantum computers process operations non-deterministically. To understand what this means, we need to know the concepts of deterministic Turing machines and non-deterministic Turing machines.

First, a deterministic Turing machine is a machine that processes a given series of commands one at a time sequentially. Common computers we use today fall into this category. Easy problems that deterministic Turing machines can solve in polynomial time, such as sorting problems, are called P problems. On the other hand, a non-deterministic Turing machine is a machine that can calculate multiple answers to a problem simultaneously, that is, a machine that finds the optimal solution among numerous possibilities. For example, in an optimal path-finding problem, when there are numerous routes from A to B, a non-deterministic Turing machine simultaneously simulates all paths to the destination and presents the path that arrives fastest as the optimal route. Problems that non-deterministic Turing machines can solve in polynomial time are called NP problems. NP problems are complex problems that require consideration of various causes and factors without standardized solutions that can be applied like formulas. Examples include optimal path finding, prime factorization, discrete logarithms, analysis of complex systems such as fluids, and natural language processing.

Now you should understand what I meant when I said earlier that quantum computers process operations non-deterministically. When conventional computers, or deterministic Turing machines that can calculate only one path at a time, try to solve NP problems, the time required increases exponentially as the complexity of the problem increases. However, for quantum computers, which are non-deterministic Turing machines, the time increases only arithmetically even as problem complexity increases. This is why people say quantum computers can easily perform calculations that conventional computers cannot. In particular, prime factorization and discrete logarithm problems constitute important parts of public key cryptographic algorithms, which is why discussions about cryptography always accompany talks about quantum computers. However, this does not mean that quantum computers are omnipotent and superior to conventional computers in all aspects. Rather, it would be more accurate to understand that conventional computers and quantum computers excel at different tasks. While quantum computers can demonstrate very powerful capabilities in certain fields, they may perform poorly depending on the type of operation. Thus, even if quantum computers become commercialized, conventional computers will still be necessary. Conventional computers will continue to be used for deterministic forms of computational work, while quantum computers will excel in solving complex problems that conventional computers struggle to process. Quantum computers and conventional computers are not in competition but rather complement each other.

With this in mind, let’s look at what quantum computers might accomplish in the future. The fields where quantum computers will excel most in the future are undoubtedly nanotechnology and data analysis. In the case of nanotechnology, quantum computers can demonstrate tremendous ability in analyzing the microscopic motion of particles. In fact, Richard Feynman first proposed the concept of quantum computers through a paper arguing that a computer based on the Schrödinger equation was needed to analyze the motion of the microscopic world. Today’s computers take a long time and lack sufficient accuracy in predicting the structure of large molecules like proteins or complex biochemical reaction processes. This is why drug development cannot rely solely on computer simulations but must go through several stages of animal testing and clinical trials. However, with quantum computers, we can predict biochemical reaction processes involving numerous interacting factors, quickly and accurately analyze various molecular structures, and use the results to accelerate the development of new drugs and materials while reducing side effects. The biggest reason drug development takes a long time is clinical trials, but with quantum computers, we could dramatically shorten the period for developing new drugs in response to new diseases like COVID-19 to just a few weeks by simplifying the clinical trial phase based on highly reliable simulations.

Quantum computers can also be useful for big data analysis. Through quantum superposition, quantum computers can quickly and accurately analyze complex and vast data with various interacting elements. Thanks to this characteristic, more accurate weather forecasts will be possible by tracking atmospheric flows and cloud movements, and they can play a crucial role in autonomous driving by identifying the movement of vehicles on roads in real-time to find optimal routes.

However, to utilize quantum computers in industry, several challenges need to be addressed. First, we need to find ways to stably implement and maintain qubits, as well as methods for quantum error correction. Since qubits easily collapse with small environmental changes, controlling them stably is a major challenge for the commercialization of quantum computers. Additionally, current quantum computers have somewhat lower computational accuracy due to quantum errors, so methods to correct these errors must be found. Various approaches such as ion traps, superconducting loops, and topological qubits are being researched for qubit implementation, each with its own advantages and disadvantages. Simultaneously, there is a need to train specialists who can write quantum algorithms and maintain, repair, and operate quantum computers. Since existing software cannot run on quantum computers, completely new types of software suitable for quantum computers will be needed.

Although AI began to receive serious attention in the 12010s, technologies that form the foundation of today’s AI, such as perceptrons, had been researched for decades. To be competitive when quantum computers receive attention like today’s AI in the future, we need to prepare from now. According to Professor Rhee June-koo of the School of Electrical and Electronic Engineering at the Korea Advanced Institute of Science and Technology (KAIST), Korea’s quantum computer technology is currently 5-10 years behind other advanced countries. Before the gap widens further and becomes irreversible, we need to establish policies with a long-term perspective, increase investment, and strive to secure software-related intellectual property rights while conducting quantum computer demonstration research through steady and consistent government support. This will require sufficient information exchange and smooth cooperation between industry, basic science researchers, and government policy makers.

References

• Hankyung Economic Dictionary, “Quantum Computer”
• IBS Science Knowledge Encyclopedia, “Much-discussed Quantum Computers, Misconceptions and Facts”
• IBS Science Knowledge Encyclopedia, “The World of Quantum Computing, Basic Science Lays the Foundation”
• Samsung Electronics Newsroom, “The World of ‘Momentary Magic’ Quantum Computers Is Coming”
• CIO Korea, “Lee Jae-yong Column | Popularization of Quantum Computers and Artificial Intelligence”
• ScienceOn, “Qubit Competition in Quantum Computers Intensifies”, http://scienceon.hani.co.kr/?document_srl=474039&mid=media&m=0, (12019 HE)
• Maeil Business Newspaper, “Quantum Computers Will Revolutionize the World in 5 Years”