Processes of applying of quantum genetic algorithm in function optimization
Download 57.19 Kb. Pdf ko'rish
|
• a parameter is placed in the input register;
• The commands that compose of function body perform some manipulation on this parameter, and then the result will be placed in the output register, and the previous state of parameter will be lost. Due to irreversibility, the last operation is impossible in quantum programming. Instead, the resulting bits will be added to the result of dividing the output register (⊗) by 2 units. In other words, XOP operation will be performed on them. This process is clearly reversible, it is enough to use a second time, and memory will be return to its initial state . MAIN PART The Quantum Genetic Algorithm (QGA) is the result of a combination of quantum computing and genetic algorithms, a new algorithm for the evolution of probability [1]. Narayanan and Moore first proposed a quantum genetic algorithm in 1996, and it has been successfully used to solve TSP problem. QGA is essentially a genetic algorithm that can be applied in an area where a traditional genetic algorithm can be applied. The efficiency of the QGA is much better than the traditional genetic algorithm. The QGA is characterized by a small population, a speed of fusion, great capabilities International Uzbekistan-Malaysia Conference on “COMPUTATIONAL MODELS AND TECHNOLOGIES (CMT2020)” AIP Conf. Proc. 2365, 030002-1–030002-5; https://doi.org/10.1063/5.0057346 Published by AIP Publishing. 978-0-7354-4110-1/$30.00 030002-1 TABLE I. TMain stages of QGA No Quantum computer Classic computer 1. Q(0) initiation of a quantum population 2. For each Q (0) → P(0), estimate and obtain the result P (0) 3. Estimate P(0) 4. while (not necessarily canceled) do 5. begin 6. t ← t + 1 7. Turn Q 8. 8. Mutation of Q 9. Q (t) → P(t), measurement transfer 10. Evaluation P(t) 11. End and reliability of global optimization. A quantum state vector has been introduced into a genetic algorithm to express the genetic code, and a quantum logic gate is used to develop chromosomes. It is possible to achieve the best results with these tools. However, there miht be some problems with the traditional QGA [2]. For instance, the traditional QGA is based on the lookup table when determining the direction of the rotation angle. In addition, a constant rotation angle negatively affects fast search and approximation [3, 4]. To solve the above problem, this article improves the quantum genetic algorithm with a flexible evolutionary process. Quantum mutations and quantum catastrophe operations are also introduced to improve the performance of the algorithm. Quantum computing is the opposite of classical computing. It uses a combination of superposition, coherence, and various quantum state qubits to perform quantum computing [5]. Quantum computing is a product of quantum mechanics usually used in the sphere of algorithms. The ability to parallelism is an important difference between quantum computing and classical computing. When calculating probabilities, the system is not in a constant state. On the contrary, it has a certain probability, and the state probability vector corresponds to various possible cases. Quantum computing is similar, the probability amplitudes of quantum states are used in quantum computing, and the probability amplitudes of quantum states are normalized, so the speed of quantum computing is √ N times higher than the classical computation speed. Quantum transformation is done through quantum revolving doors. Quantum computing has some peculiarities compared to classical computing [6]. Some mechanisms of these functions will become possible when the optimization algorithms are implemented to improve the traditional optimization algorithm. Our main goal in the framework of this article is to consider the canonical classification of quantum evolution algorithms, the process of solving problems using quantum genetic algorithms. In general, the quantum genetic algorithm includes the main steps listed in Table 1 [7, 8]. Quantum genetic algorithms might be revised as classical optimization methods based on quantum computational principles. Programs implementing such methods might be run on a digital computer without practical or theoretical difficulties. However, one of the problems of quantum artificial intelligence today is the development of real algorithms and programs for quantum evolution that can be executed on a quantum computer. Also, some problems arise when we translate the main steps of a simple genetic algorithm into a quantum version. This is unexpected because a simple genetic algorithm is similar to the quantum Grover algorithm, simple genetic algorithms are considered parallel search methods. One of the main problems in quantum genetic algorithms is to determine the population state of individuals without disturbing the superposition state of these chromosomes [7]. In addition, one of the most important problems that remains unresolved today is how to implement the crossover operator on a quantum computer. Given that mutations can be easily performed on a quantum computer, i.e., using the Pauli (x) gate, it is unclear how to perform a crossover using quantum mechanical phenomena for this purpose [9, 10]. One of the more interesting ideas came from 2006, when the first steps has been done to implement a genetic algorithm in a quantum computer. The authors of these studies proposed a true quantum evolution algorithm called the abbreviated quantum genetic algorithm (AQGA). This algorithm consists of the main steps described below (Table 2). The above algorithm begins, first of all, by creating a superposition of all individuals, that is, N or Q(t) chromosomes of the population, i.e. Download 57.19 Kb. Do'stlaringiz bilan baham: |
Ma'lumotlar bazasi mualliflik huquqi bilan himoyalangan ©fayllar.org 2024
ma'muriyatiga murojaat qiling
ma'muriyatiga murojaat qiling