الفهرس 
IntroductionOnly 14 pages are availabe for public view
 |  | from | 198 |  |  |
| | | from | 198 | | |
Abstract
Objectives
Study single- and multi-objective meta-inference algorithms to solve optimization problems. Application of hyper inference to solve optimization problems. Review of reinforcement methods for solving overdetermination Algorithm issues.
Methodology
To improve the efficiency of the optimization algorithm, there are two main processes of optimization, exploration (diversification) and exploitation (condensation), where the exploration optimization algorithm searches for approximate solutions in the broad search area of the problem, so that it gets the needs of the solutions to be refined. This process of exploring the search area helps to avoid trapping at local minimum points above the search area; On the other hand, the exploit optimization algorithm is looking for a solution in a limited area of the search space, and trying to optimize the most promising solutions explored in the previous stage. The optimization algorithm must be able to balance them, since the meta-inference is considered problem-dependent, necessitating the optimization algorithm so that it is self-adaptive to the optimization parameters.
Results
In this thesis, a hybrid optimization algorithm is proposed to solve the problems of engineering and synthesis optimization; Also validated in the post of the CEC’2017 standard. Furthermore, the quantum cloning circuit was formulated for the optimization problem, and it was optimized with optimization algorithms to bring the quantitative reproduction error rate down to 10E - 8. Finally, the multipurpose descriptive algorithms were revised, in addition to a multi-objective MOSMA validation in CEC’2020. Standard functions. Apply metadata optimization algorithms to solve the real world. Optimization of quantum cloning circuit parameters using an AGDE meta optimization algorithm. First attempt to use optimization techniques to solve this problem. Achieved cloning in quantum computing with an error rate of 10E - 8. An optimized single-optimized hybrid algorithm SMA-AGDE to solve combinatorial and engineering optimization problems. Also introduced as SMA-AGDE certified in the functions of the Standard Test group CEC’2017. The multipurpose MOSMA optimization algorithm has been validated in the CEC’2020 test group with six known multi-target algorithms.
Recommendations:
According to the results of this study, the proposed SMA-AGDE algorithm can be applied to other process optimization problems such as control engineering where multiple criteria conflict, and more practical optimization problems such as classification and prediction problems as well as multi-objective problems. The multipurpose MOSMA algorithm can also be used to improve applications related to machine learning, such as feature selection, data pre-processing, and parameter optimization.