Quantum computing
Quantum computing has the potential to be the next disruptive technology, with numerous potential uses and effects on markets and businesses.Data is represented and processed by quantum computers through the application of quantum mechanics principles like superposition and entanglement.Quantum computers can solve very specific, complex problems much more quickly than conventional computers thanks to these two guiding principles.This introduction provides a brief overview of the three layers of a quantum computer against this background:the application layer, system software, and hardware.In addition, we discuss potential research directions in the field of information systems and potential fields of application for quantum computing.
Introduction Quantum computing looks like the next disruptive technology with a lot of potential uses and repercussions for businesses and markets.According to a recent McKinsey report (Hazan et al.,), the financial, chemical, pharmaceutical, and automotive industries will dominate the global market for quantum computing by 2035.2020).Today, the largest technology companies in the world, including Google, IBM, Microsoft, Amazon, and Alibaba, are already investing billions of dollars in the research and development of their quantum computers. Through cloud infrastructures, they provide the general public with partial access to these quantum computers.However, not only do industry players not invest, but governments also do so. For instance, China invested $10 billion in the national quantum computing laboratory, the US government provided $1 billion, and the EU budget totals more than $1 billion (Castellvecchi).2018;(2018) (Deicker and Yasiejko).
Data is represented and manipulated by quantum computers using quantum mechanics principles like superposition and entanglement.Quantum computers can solve very specific, complex problems much more quickly than conventional computers thanks to these two guiding principles.Interference also plays a significant role, particularly when reading data from a quantum computer.Instead of working sequentially, quantum computers are able to compute and test enormous combinations of hypotheses simultaneously.Furthermore, some quantum algorithms can be designed to solve problems in significantly fewer steps than their classical counterparts (which are simpler).In this manner, quantum processing could be a significant leap forward in present day IT before long and start the change to the "fifth modern upset".
The initial experiments show promising results, such as Google's 2019 experiment, in which the company asserts to have achieved a so-called quantum advantage (also known as IBM's "quantum advantage").They were able to demonstrate, through an artificial experiment, that a programmable quantum device could solve a problem that a classical computer could not in a time-appropriate manner.However, Google's quantum computer's solution to the problem was tailored to the quantum hardware used, and it has no real-world applications.In any case, it served as a significant proof of concept.Also in 2020, Chinese researchers said they had developed a quantum computer that could do some calculations 100 trillion times faster than the most advanced supercomputer in the world.
Experts anticipate that quantum computing will provide unprecedented benefits, particularly in the areas of optimization, artificial intelligence, and simulation, given its current stage of development.The chemical and pharmaceutical industries' molecular simulations are likely to be the first real applications of quantum computers.This is because molecules follow the laws of quantum mechanics directly, so simulating them with quantum computers is the most natural thing to do.The financial sector, transportation and logistics, the global energy and materials sector, as well as fields like meteorology and cyber security, may soon benefit.From hardware architecture and data management to application software and algorithms, quantum computing currently faces enormous unsolved challenges in physics and computer science that necessitate fundamental research in all of these areas and beyond.
This Foundation covers the fundamentals of quantum computing and outlines research opportunities to support IS research.As a result, in the second section, we provide a brief description of the quantum computer system and its three layers:application layer, system software, and hardware.The potential fields of application for quantum computing are discussed in the third section.
Footnote We discuss each layer's potential research opportunities in the context of electronic markets in detail in light of these and the adopted conceptual layer view of quantum computing.Around quantum computing technology itself, a brand-new ecosystem is already forming, which raises concerns regarding innovations in business models and processes, IT challenges, and the acquisition of start-ups and key suppliers.such as Alibaba, IBM, Google, or individual development
QUANTUM COMPUTING SYSTEM
In 1980, Paul Benioff proposed the theoretical concept of a quantum computer, also known as a quantum touring machine.Richard Feynman proposed the first real-world use of a quantum computer in 1982:efficient quantum system simulationsA quantum computer can be described as a universal computing device that uses quantum mechanics-derived properties to transform quantum bits (or qubits) and store information within them.Importantly, the quantum computer does not intend to become a general-purpose computer that operates on its own. Instead, it performs quantum computing, a type of computation that collects the various states of qubits, such as superposition, interference, and entanglement, to perform calculations.They will be profoundly particular gadgets that can tackle explicit undertakings a lot quicker than old style figuring.For loading input and output data, retrieving computation results, and controlling the electronic and internal processes of the quantum computer, operating a classical computer will absolutely be necessary.
As a result, quantum computers and conventional computers make up a quantum computing system that makes it possible for quantum computers to carry out quantum computations.We use Ding and Chong's model to show the various layers of a quantum computing system for three reasons.To begin, it enables us to demonstrate the fundamental mechanisms and elements of a quantum component system by analytically distinguishing the key components.Second, it is based on an analytics distinction between hardware, system software, and application. This distinction is reflected in conceptual views of computing architectures, such as cloud computing (Infrastructure-as-a-Service, Platform-as-a-Service, and Software-as-a-Service) or digital technologies' layered modular architecture (Yoo et al.,2010).Thirdly, our expert informants explained the state of the art, the challenges facing organizations today, and the operation of quantum computing systems by distinguishing between similar layers in their interview statements.Figure 1 depicts a quantum computing system with a van Neumann architecture for classical computing and a three-layer architecture for a quantum computer, which we will discuss in turn.
Hardware layer The way information is stored is one of the main differences between quantum and classical computers.Quantum computers use quantum bits (or qubits), which can simultaneously hold any linear combination of zero and one, as opposed to classical computers, which use bits, which can have the value of zero or one.Qubits benefit from the properties of quantum mechanics, particularly the effect of superposition (see Fig. for a visual).2).
Superposition
Freely talking, a qubit is portrayed by its likelihood of being either zero or one and not by the particular worth of nothing or one.As a result, a qubit may have a probability of 60% of zero and 40% of one.Importantly, the qubit only "collapses" to the single classically defined value of zero or one when measuring its state.The advantage of superposition is that a quantum computer with just four qubits can simultaneously represent 16 numbers with four digits.While a conventional computer can only represent a single four-digit number with a sequence of four bits, the number of representable states doubles with each additional qubit.
The ability to perform an exponential number of calculations simultaneously is the real benefit of quantum computing.Although it is possible to read only the solution to a single calculation at the end of every program, it is possible to create a quantum algorithm that increases the likelihood that the desired outcome will be achieved.For instance, we might be attempting to determine whether there is any rare turbulence that could lead to a plane crash.We could simply test almost all possible air conditions simultaneously on a quantum computer and only read out the result that causes the plane to crash, as opposed to simulating billions of combinations of air conditions on a classical computer and checking their individual results.
Entanglement Quantum computing is not just about qubits.Quantum mechanics
also has the property of entanglement.Entanglement occurs when one qubit's state is dependent on another qubit's state.As a result, when two qubits are entangled, any flip or rotation made to one qubit would also affect the other qubit.In addition, the state of both qubits collapses to either one or zero (depending on their probabilities) when either one of the qubits' states is measured.Even when the qubits are in close proximity to one another, this holds true.As a result, the advantage of entanglement is that when a qubit influences other qubits in its vicinity, they are all working together to solve a problem.
As a result, qubits can be correlated in a way that bits in conventional computers cannot.As a result, the quantum computer will be able to process information in a fundamentally different manner than a conventional computer, opening up new possibilities.Superdense coding, in which one entangled qubit is used to transport two conventional bits of information, is one illustration.For secure quantum key distribution, this process is particularly intriguing.Quantum entanglement and other quantum phenomena are used to implement a cryptographic protocol in this secure method of communication.It makes it possible for two parties to generate a shared random secret key (entangled qubit) that is only known to them. This key can be used to encrypt and decrypt messages.
We now talk about ways to physically represent and manipulate qubits, based on the fundamentals of quantum mechanics.The approaches can broadly be divided into two main groups:digital gate-based quantum computing and analog quantum computing, respectively.
We now talk about ways to physically represent and manipulate qubits, based on the fundamentals of quantum mechanics.The approaches can broadly be divided into two main groups:digital gate-based quantum computing and analog quantum computing, respectively.
Analog quantum computing
In analog quantum computing, quantum operations change the quantum state in such a way that the information encoded in the final system is highly likely to match the desired response.A type of universal quantum computing called adiabatic quantum computers is an illustration of analog quantum computing.According to Vinci & Lidar (2017), quantum annealing is a distinct type of adiabatic quantum computers. It is a framework that incorporates algorithms and hardware designed to solve computational issues through quantum evolution toward the ground states.Quantum annealing takes advantage of the fact that physical systems strive for the lowest energy state, such as when hot things cool down over time or when objects roll downhill.Therefore, the energetically most advantageous state in quantum annealing corresponds to the optimization problem's solution.The quantum annealer is able to calculate all possible solutions simultaneously thanks to the property of superposition. This significantly speeds up the calculation process when compared to traditional computers (Shin et al.,2014).Companies like D-Wave use quantum annealing because it works best for optimization problems or probabilistic sampling.However, the quantum annealing method's likelihood of ever achieving a significant quantum speedup is currently unknown.
Digital gate-based quantum computing Digital gates are used to manipulate the information encoded in qubits in digital gate-based quantum computing.In digital gate-based quantum computers, the evolution of quantum states is manipulated in terms of activity and controlled to find the optimal solution, in contrast to the analog approach, in which you sample the natural evolution of quantum states to find the optimal low-energy state.In contrast to quantum annealing, the qubit state can be used to solve a wide range of problems because it is actively manipulated, which makes it much more adaptable.The idea behind digital gate-based quantum computing is very similar to that of classical computation.A computer runs a classical algorithm as a series of instructions (gates like AND, OR, NOT,...).They follow a set of rules to flip classical bits between zero and one states, either individually or in pairs.By rotating and shifting qubits between various superpositions of the zero and one states, as well as between various entangled states, quantum gates can operate directly on one or more qubits.IBM, Google, and Rigetti are just a few of the companies that use digital gate-based quantum computing.
System software layer The system software layer orchestrates the system's processes in order to take advantage of the qubits' potentials (superposition and entanglement). It builds on top of the hardware layer.The challenges posed by quantum states that are thermodynamically unstable must be addressed by this layer.It corrects errors and actively reduces thermal noise within and around the quantum system.
There are numerous potential sources of noise in quantum computing.Digital gate-based quantum computers, in particular, are extremely sensitive to environmental changes like vibration and temperature changes.Inadequate control over the quantum hardware or manufacturing flaws can also result in noise.The chips of the majority of quantum computers must even be cooled to one hundredth of a degree above absolute zero in order to function.Consequently, since commotion can't be kept away from, the principal time of quantum PCs is likewise called loud Middle Scale Quantum PC (NISQ, Preskill, 2018).This abbreviation implies that before we can construct useful quantum computers with hundreds or even thousands of usable qubits, the error rates of the existing quantum hardware, which consists of dozens of qubits, must be reduced.
Qubit decoherence, in which environmental influences cause quantum states to change at random, can be caused by noise in the environment.This is a problem because, unless the error is fixed during the calculation, a single calculation error typically leads to an incorrect result.Error correction is essential due to the fact that it is impossible to eliminate all types of noise.The goal of ongoing research on quantum error correction is to achieve fault tolerance at the system level.Quantum error correction distinguishes between qubits that are logical and physical.A collection of physical qubits, which are required for error correction, represent logical qubits.Physical qubits collaborate to correct individual qubit-specific errors.A single physical qubit is more likely to make a calculation error than a group of physical qubits.Sadly, mechanisms that correct errors can also make mistakes.For one almost error-free logical qubit, the relation typically entails between five and nine physical qubits, depending on the error-correcting mechanism.
This can be accomplished in a number of ways, one of which is by representing each qubit as a group of several physical qubits that, loosely speaking, collaborate to correct errors on individual qubits.Because it does not need to be corrected for errors, a perfect physical qubit can function as a logical qubit.Today, scaling up to thousands of qubits presents the greatest obstacle.Due to high error rates, this advantage, which doubles the amount of available computational space for calculations with each qubit, cannot be utilized to its full potential at this time.IBM is a well-known company that has set its sights on exceeding 1,000 qubits by 2023, despite the fact that machines with 60–100 qubits are currently available.
Application layer
The unsolved issue of effective quantum memory is one of the main obstacles facing today's quantum computers.Quantum random access memory, or QRAM, is the subject of several theoretical proposals.Recent research has shown that there are a number of possible routes to take, despite the fact that it may be experimentally challenging to construct (just like the quantum computer itself).As a result, there is currently no effective method for long-term memory storage of qubit states for use in other calculations.
As a result, before the qubits lose their information, data must be transferred from a classical computer to the intended quantum computer and states must be read (measured) by the classical computer following the calculation.Quantum states can't be copied and used in calculations either because of the no cloning theorem.A SWAP operation is the only method that can be used to remove a quantum state from memory and load it into a quantum program.
A quantum state collapses to either one or zero when measured.In this way, we have not a chance of figuring out what express a certain qubit is in.Having multiple copies of the same qubit and measuring them all is the only way we can roughly determine its state.
Quantum algorithms may not be able to speed up the entire algorithm at the macro level because reading classical data may dominate the cost.It may not be possible to precisely read the data because it does not meet the computing requirements for some tasks.This is especially true for artificial intelligence and machine learning techniques that require large data sets.
Creating an algorithm that maximizes the probability of measuring the desired outcome is the most important step in developing one that can be used with quantum computers.We typically only care about a small subset of the quantum computer's output, which may be an exponentially large number of solutions.The art of building quantum algorithms is finding them without repeatedly running the entire algorithm.
Three of the most significant quantum algorithms are listed here.
The quantum search algorithm is another name for Grover's algorithm.When searching an unordered list or unstructured database, Grover's algorithm is used.Traditionally, in order to locate a particular item in a database of size N, we must search through N/2 items on average to locate it.This can be accomplished in approximately N steps using Grover's algorithm.This may be significantly faster for a large N.The term for this is a quadratic speedup.
The quantum search algorithm is another name for Grover's algorithm.When searching an unordered list or unstructured database, Grover's algorithm is used.Traditionally, in order to locate a particular item in a database of size N, we must search through N/2 items on average to locate it.This can be accomplished in approximately N steps using Grover's algorithm.This may be significantly faster for a large N.The term for this is a quadratic speedup.
The integer factorization algorithm, also known as Shor's algorithm, can factorize integers nearly exponentially faster than the fastest classical algorithm that is currently known.Since factoring integers is a computationally challenging process, it serves as the foundation for RSA encryption.
The quantum algorithm for linear systems of equations, also known as HHL (Harrow Hassidim Lloyd), is also used.A linear system's solution x can be estimated using the algorithm (Ax = b), where A is a vector and b is a matrix.
A growing number of established commercial businesses are investing in quantum technology as a result of the significant advancements in hardware.Examples include Daimler, which announced progress in the field of materials research, Boehringer Ingelheim, which recently announced a research partnership with Google, and BASF, a chemistry giant that strives to remain at the forefront of chemistry research and business.
To address the computational issues of today that current computers are unable to or only partially address, quantum computing has three essential capabilities that are advantageous to businesses:1) algebraic, 3) simulation, and 4) search and graph.
These abilities decide the likely utilizations of this innovation in various enterprises, like money, science and pharma, assembling, energy, or network protection .The various problem types, approaches, and potential use cases are summarized in Table 1.
Search and graph The ability of a qubit to theoretically represent an infinite number of states makes it possible to solve difficult combinatorial optimization problems. This is currently one of the main areas where quantum computing technologies like D-Wave's solution can be used.
The process of locating one or more optimal solutions to a problem is known as combinatorial optimization. Supply chain optimization, optimizing public transportation schedules and routes, package delivery, and other such issues are examples of such issues.In a discrete (finite) but very large configuration space (a set of states), these solutions are sought.The objective is to find the best solution to optimize the objective function, and the set of possible solutions can be defined with a number of constraints.
With classical computers, it is extremely difficult to find the best or even a single good solution to certain complex problems in a reasonable amount of time or with sufficient accuracy.
This is due to the large problem spaces of these problems. Combinatorial optimization problems of this kind frequently present a significant obstacle for both the public and private sectors.
They are often easy to explain, but they are actually very hard to solve. There are subclasses of the order, assignment, grouping, and selection classes of combinatorial optimization problems, such as the knapsack and traveling salesman problems. There is no known algorithm that can easily compute these problems directly, despite the fact that there may be numerous qualitatively distinct solutions to a problem. Searching extremely large problem spaces takes a lot of time and computing power.
Quantum computers are anticipated to have a significant impact on the financial services sector. Players who specialize in portfolio optimization and arbitrage may gain from this. A subset of the existing financial instruments should be chosen to achieve a certain portfolio volume while simultaneously considering a large number of factors to minimize risk and achieve profitability.
Quantum computers are anticipated to have a significant impact on the financial services sector. Players who specialize in portfolio optimization and arbitrage may gain from this. A subset of the existing financial instruments should be chosen to achieve a certain portfolio volume while simultaneously considering a large number of factors to minimize risk and achieve profitability.
In addition, Deutsche Börse, a German company that provides marketplace organizing for the trading of shares, has already experimented with the applicability of quantum computing for a sensitivity analysis on one of their risk models. This is a computation that would be too costly to run on classical computers. However, it can be done with quantum computing.
The optimization of flow, such as that of goods or traffic, is another use for quantum computing because it can be used to solve optimization problems.VW has already demonstrated in a pilot project with D-Wave Systems how to utilize quantum annealing technologies to simplify traffic flow in the city of Lisbon. The project began in late 2016 with a proof-of-concept project.
By developing a traffic-flow optimization program that utilized the GPS coordinates of 418 taxis in Beijing to alleviate traffic congestion, it investigated the readiness of quantum computing.
In addition, certain prime factorization procedures that are crucial to the safe encryption of data make quantum computers superior to classical ones.The Shor algorithm, which divides a number into its prime factors and is frequently utilized in cryptography and cybersecurity, is a well-known illustration of this.Classical computer technology would be unable to decrypt a quantum-encrypted dataset, at least not in time periods relevant to human users.On the other hand, a quantum computer could easily decrypt data encrypted using conventional RSA technology, a phenomenon that could be referred to as a "quantum threat."
Algebraic
For narrow AI approaches, quantum computing's capacity to accelerate optimization problems is crucial. For machine learning and artificial intelligence, complex network architectures and weights can be calculated with the assistance of quantum computing. Large metric transformation and calculation demonstrate quantum computing's advantages. In the context of supervised learning, for instance, the model aims to minimize the error between the model's prediction and the input and the provided adequate output or label.
This kind of problem can be solved in a variety of ways with quantum computers, which also speed up calculation and allow for more complex network architectures. According to an experiment conducted by Cambridge Quantum Computing, they may be applicable to all relevant practices or subtasks of artificial intelligence, such as image processing, computer vision, and natural language processing.Having said that, it is essential to keep in mind that there has not yet been discovered a machine learning algorithm with a demonstrable speedup.
Simulation
A fundamental advantage of a quantum computer over conventional computers is:It is capable of simulating other quantum systems, such as a nitrogen molecule, much more effectively than any current computer system.For old style PCs, even particles with similarly low intricacy address a practically unsolvable errand.
Richard Feynman proved theoretically in the 1980s that a quantum-based computer could simulate molecules.Since then, researchers have attempted to simulate the molecule's quantum system by transferring it into the quantum computer or another quantum system.
The simulation of more effective catalysts for the Haber–Bosch process, which currently accounts for about 1% to 2% of global energy consumption, is one promising new application of quantum computers.Better catalysts might help slow down global warming by using less energy.
Simulations on classical computers may not be as suitable for this application as quantum computers without full error correction.
In addition, the process of developing drugs and active ingredients frequently takes a very long time and is very expensive.This is especially because a lot of substances need to be tried out in the real world through trial and error.However, quantum computing may be able to virtually replicate the behaviors based on the same principles as quantum physics. As a result, simulation-based research may eventually replace this costly procedure.
In light of BASF's stringent requirements for quantum chemical calculations' accuracy, for instance, the company collaborated with HQS to investigate the potential application of quantum computing.
They were particularly interested in comprehending the quantum mechanical calculation of the energy course of chemical reactions, as this actually enables the prediction of the probable course (i.e., which products, by-products, etc.,are formed, how can I use catalysts to speed up the reaction, etc.)of chemical processesThe limitations of conventional computing approaches are reached by this application of necessary methods.
Material research on how batteries work is also thought to be important for today's electromobility and is already being pursued by automakers like Volkswagen.
Link to Information Systems The majority of expert estimates still place the widespread industrial application of quantum computing at least five to ten years away, despite the substantial investments in the field.
It is still unclear how exactly it manifests in many important areas.Therefore, it is the responsibility of the current research community to imaginatively imagine and investigate quantum computing's full potential as well as its sociotechnological consequences.
We propose the following four initial directions for research on quantum computing in information systems based on an analysis of the existing literature and interviews with 21 leading industry and research experts:1) quantum computing ecosystems as a new networked business; 2) digital understanding as the basis for use cases and ecosystems; 3) quantum computing as a challenge for IT organizations and IT service providers; and 4) the skills required to utilize quantum computing in the quantum computing field.
All of these directions try to take into account the fact that, despite its potential to disrupt the electronic market, ecosystem, and its participants, quantum computing will initially be an extension of computing capabilities (see Fig.3), while brand-new ecosystem participants like IonQ and Rigetti are already establishing themselves.
0 Comments