1. Why might quantum computing become an outlet?
Quantum computing is a new type of computing that uses the principles of quantum mechanics for computing. Unlike traditional classical computers, quantum computers use quantum bits (qubits) for information processing. Since qubits can be in a superposition state, this allows quantum computers to perform certain types of calculations (such as integer decomposition, searching unsorted databases, When simulating the quantum behavior of matter, etc.), it has a huge advantage over classical computers.
In classical computing, the storage and processing of data is based on bits, and the value range of each bit is the set {0,1}. That is to say, the data value stored in a register can only be 0 and 1. But after we understand wave-particle duality, we can extend this concept to computer systems. If there is a bit that can have two contradictory information of 1 and 0 at the same time, then the computational efficiency will increase exponentially. Let's think about a simple case. When calculating the simple mathematical logic of 9+5=14, it needs at least two 4-bit registers to store the two data of 1001 and 0101 respectively, and then make them Accumulate, and because the result of the calculation is 14 (binary representation 1110). After this calculation is over, if I want to calculate 10+5, at this time, in the logic of the classical computer, the values of the original two registers need to be updated to 1010 and 0101. Although there is no updated data in the second register, it is logically necessary to check the value of the register, and checking this will also consume time and power (of course, in this example, this overhead can be ignored). However, suppose we mentioned a bit that can carry both 0 and 1 information at the same time, we denote it by Q. When calculating the two mathematical operations of 9+5 and 10+5, the values in the two registers are both QQQQ. That is to say, when the computer does not expand the register by carry, QQQQ can express a total of 16 values from 1 to 15. In a classical computer, the value of the register corresponds to the decimal number one by one, and we must retrieve and update it before proceeding to the next calculation. The bit Q we assume is a qubit, which can carry both 1 and 0 information at the same time. Therefore, from the perspective of a computer, the efficiency of a qubit is exponential for the number of bits of the same scale. The development of quantum computing could lead to a computing revolution, because its theoretical computing power far exceeds that of existing classical computers. Quantum computing can provide exponential speedup on certain problems. For example, Shor's algorithm can solve integer factorization problems that classical computers need exponential time to solve in polynomial time. If the development of quantum computing comes to fruition, it could potentially change the way we deal with complex problems and big data, with major implications in science, engineering, finance, and many other fields.
However, at the current stage, quantum computing is still in the theoretical research stage, and its actual deployment requires a high degree of technology integration in multiple fields, not only in the computer field, but also in many high-tech industries including environmental engineering, material engineering, physics, and mathematics. The quantum method currently deployed and proposed has theoretically verified the expressiveness of a certain field that has the potential far beyond classical computing, but it is still in the laboratory stage in essence, and there is still a short distance from commercialization. A representative example is quantum machine learning. Quantum machine learning methods provide exponentially higher efficiency than classical machine learning. However, when enterprises actually deploy or run data experiments in laboratories, the results obtained are not only far lower than expected, It may not even be as good as the excellent algorithms in classical machine learning, but the time-effective benefits are higher.
2. Current applications of quantum computing in various fields
Quantum computing is currently in its infancy, but its potential applications in several fields are compelling. The following are a few examples (please note that these are examples of potential applications. There are commercial structures for deploying quantum computing in the actual industry chain, but they are very few, mainly in the scientific research stage, and commercialization is far away):
Scientific research : Physicists and chemists can use quantum computers to simulate and study complex quantum systems, such as high-temperature superconductors and complex chemical reactions, which are difficult on conventional computers. In particular, physics based on machine learning can help physicists better understand the quantum world.
Machine Learning and Artificial Intelligence : Due to their parallel processing capabilities, quantum computers may be able to outperform classical computers on certain types of machine learning tasks, such as optimization and clustering. However, quantum computers are not unconditionally faster than classical computers. The speed of quantum computing is specific, and it only has a very high advantage in a specific field. However, it still does not have quantum computing that can perform universal operations like classical computing. computer. It still takes more than ten years of technical precipitation for the masses and large enterprises to deploy a quantum computer.
Cryptography and information security : Shor’s algorithm can crack most of the existing public key cryptosystems on quantum computers, which poses a threat to information security, but it also gave birth to the development of quantum cryptography, which can theoretically Provides unconditionally secure communication.
Drug discovery and biotechnology : Quantum computers have the potential to simulate complex biomolecules and chemical reactions, which has significant potential value for the fields of drug discovery and biotechnology. For example, they can help researchers discover new drugs faster and improve the efficiency and effectiveness of biotechnology.
Physics and materials science : For physicists and materials scientists, quantum computers may become a powerful tool to help them simulate and understand complex quantum systems, such as high-temperature superconductors, new energy materials, etc.
Financial and economic modeling : In finance, quantum computers may help solve complex optimization problems such as portfolio optimization, risk management, and pricing.
Climate Science : Modeling the global climate system requires a lot of computing power, and it also requires processing huge amounts of data. The parallel processing power of quantum computers may make such simulations more efficient, helping us better understand global climate change and formulate corresponding policies.
Supply chain and logistics : In supply chain and logistics, there are many complex optimization problems, such as how to allocate resources most efficiently, or how to plan optimal transportation routes. These problems are intractable on classical computers, but quantum computers may offer a solution.
Space exploration and astrophysics : Quantum computers can simulate and analyze complex celestial phenomena, such as black holes, reactions inside stars, etc. In addition, quantum communication technology may also improve deep space communication and provide support for future space exploration.
Next is information on some companies that actually deploy quantum computing commercially. Although these companies provide quantum computing commercialization, they are more about providing technical support for users to access quantum computers, rather than deploying quantum computing application products.
- IBM: IBM was one of the early leaders in quantum computing, developing a range of quantum hardware and providing access to that hardware through the cloud. IBM also offers an open-source quantum programming framework called Qiskit.
- Google: Google's quantum computing project is also very active, and they claimed to have achieved "quantum superiority" in 2019. This means their quantum computer outperforms the strongest classical computers at certain tasks.
- Microsoft: Microsoft is developing a state-of-the-art qubit technology called Station Q. In addition, they also developed a quantum programming language Q#, and provided a quantum computing cloud service called - Azure Quantum.
- D-Wave Systems: This is a Canadian company focused on developing quantum annealing technology, a special quantum computing method that is especially suitable for solving optimization problems.
- Rigetti Computing: This is a start-up company based in the United States. They are developing their own superconducting quantum computer and provide a quantum programming platform called Forest.
- Volkswagen: German automaker Volkswagen has teamed up with quantum computing company D-Wave Systems to conduct research on traffic flow optimization on quantum computers.
- JP Morgan Chase: The financial institution is working with IBM on the use of quantum algorithms for financial modeling and risk analysis.
- ExxonMobil: The oil and gas company is working with IBM to investigate the use of quantum computing in materials science and optimization problems to improve the efficiency of its industrial processes.
3. Challenges faced by quantum computing
Difficulty upgrading the technical structure : It is extremely challenging to build and maintain a stable quantum computer. Qubits are very susceptible to environmental influences and lose their quantum properties, which is known as "quantum decoherence". In addition, the current error correction technology is not mature enough, which is also critical for the reliable operation of quantum computers.
ROI : Quantum computers are very expensive to manufacture and maintain. A quantum computer needs to run at extremely low temperatures, which requires a lot of cooling equipment, making quantum computers much more expensive than conventional computers.
Impact on modern security systems : While quantum computing may bring new security guarantees, it may also disrupt existing security frameworks. For example, Shor's algorithm can crack encryption algorithms such as RSA on quantum computers, which may pose a threat to existing information security.
Quantum computers undoubtedly have great potential to realize high-speed technological integration in smart cities. However, the huge potential of quantum computers also has corresponding limitations. The limitations are mainly as follows:
Quantum entanglement : Only when qubits are entangled with each other can they have exponential computing power. The state of any qubit needs to be correlated with other qubits. In order to form such a connection between two qubits, whether it is photons, electrons or other quanta, an intermediate quantum system is needed to directly or indirectly help them interact. Thus interacting at some point with each qubit that needs to be entangled. In the realization of quantum computing, this long-distance interaction will consume some qubits of the calculation, and the useful qubits will be reduced. To eliminate this overhead, some two-qubit operations in general-purpose gate sets require multiple executions of the basic gate operation, which is especially significant when qubit and gate operations are constrained.
no cloning theorem: Although it is possible to switch the state of one set of qubits to another set of qubits, this information in the basic qubit is deleted. That is, the duplication of the quantum system cannot be realized, and the copy of the intermediate state or partial result generated and stored in the memory is the way we want, but due to this reason, the effective storage of the intermediate data cannot be realized. Quantum algorithms need a way to access the classical bits of storage, so that it is clear which bits are being queried and loaded into quantum memory.
Noise : Like classical logic gates, basic gate operations cannot eliminate bit errors in the input signal and the gate operation process, which will accumulate over time and affect the accuracy of the calculation. When there are enough bit errors, it will cause measurement errors and even decoherence.
After the computation is complete, it is not possible to actually observe the complete state of the computer. Under normal circumstances, there are n possible outputs and the probability of obtaining each probability is equal. We can manually intervene in the coefficients of these states for the desired output, so that the probability of extracting the desired state is greatly increased when we measure the results, but doing so will also reduce the overall speed of quantum operations.
Improvement of quantum computing capabilities : Quantum cryptography requires a large number of quantum states, and the current quantum computing capabilities are still not perfect. The research on the sustainability of quantum entanglement, the research on the expansion of qubits, the development of quantum algorithms, and the maintenance of coherence all need continuous exploration.
Anti-interference : Particles in quantum cryptography are easily disturbed and leave the quantum state, so it is necessary to improve the resistance to interference. Technologies that can improve the robustness of qubits and have more precise control over the parameters of the environment remain unresolved challenges.
Security of Quantum Channels: The information in quantum cryptography is transmitted on the quantum channel, and the security of the channel is an important premise of quantum cryptography. In addition, although the QKD technology can allow both communication parties to find out whether the communication has been wiretapped, it cannot prevent wiretapping or the fact that information has been leaked. Therefore, it is still inappropriate to describe QKD technology as absolutely safe, and other ways are needed to achieve more secure quantum communication. When eavesdropping occurs, the quantum communication is interrupted, the attacker cannot observe the communication content and the two parties in the communication can clearly know that there is eavesdropping on the communication, but this scheme also has disadvantages. When eavesdropping occurs frequently, the frequent interruption of communication will lead to communication efficiency. extremely low. In conclusion, research on secure quantum communication channels is still a huge topic.
While great progress has been made in the field of quantum technology, universal error-correcting quantum computers with a limited number of qubits are far from being realized. It is not yet clear how many logical quantums a quantum computer needs to surpass a classical computer. Different algorithms may produce different ratios according to different characteristics, so this increases the difficulty for the realization of QML. The difficulty of implementing quantum machine learning now appears to be a very large state preparation problem. Any state preparation is exponential in the number of qubits in the discrete gate set, which provides a limit for the performance of all algorithms and limits the state of the algorithm when it is initialized. limited.
4. The Future Development Trend of Quantum Computing
At this stage, quantum computing belongs to the era of vigorous development, and quantum machine learning is one of the main directions of future development. It is a technology that can help us improve computing power and become the next step for effective and meaningful analysis of these massive data sets. . Professor Seth Lloyd of the Massachusetts Institute of Technology (MIT) proposed theoretical predictions that using the parallel computing advantages of quantum systems in processing high-dimensional vectors can bring exponential acceleration to classical technologies such as machine learning. It will be able to far exceed the computing speed of existing classical computers.
Although quantum computing faces many challenges, its development potential has led many scientific research institutions and companies, including Google, IBM, Microsoft, etc., to actively invest in research and development. In the future, we may see the following development trends:
Advances in hardware : With the deepening of scientific research and technological progress, we may see more stable, more reliable, and larger-scale quantum computers.
Development of quantum software and algorithms : At present, we already have some quantum algorithms, such as Shor's algorithm and Grover's algorithm. In the future, we may see the emergence of more quantum algorithms, as well as the development of more mature quantum programming languages and tools.
Realization of commercial applications : As quantum computing technology matures, we may see its practical application in more fields, such as drug design, logistics optimization, financial risk assessment, etc.
In general, although quantum computing faces many challenges, its huge potential makes it an important direction in scientific research and industry. As our understanding of the quantum world improves, we have reason to believe that quantum computing will play an increasingly important role in the future.