To address the challenge of high computational complexity in determining the cryptographic characteristics of S-boxes created through genetic algorithms, a novel method for S-box design is introduced that incorporates...
详细信息
genetic algorithms (GAs) are a powerful class of optimization techniques inspired by the principles of natural selection and genetics. One of the theoretical cornerstones of GAs is schema theory, which provides a fram...
详细信息
We propose a novel methodology integrating genetic algorithms (GAs) with quantitative back-testing to optimize a multilayer perceptron (MLP) model for stock price prediction and trading strategies. Addressing three ma...
详细信息
This study develops an advanced dynamic optimization model that leverages Mixed-Integer Linear Programming (MILP) alongside the genetic Algorithm (GA) to tackle complex resource allocation and configuration challenges...
详细信息
ISBN:
(数字)9798331536169
ISBN:
(纸本)9798331536176
This study develops an advanced dynamic optimization model that leverages Mixed-Integer Linear Programming (MILP) alongside the genetic Algorithm (GA) to tackle complex resource allocation and configuration challenges under a multitude of constraints. By incorporating uncertain variables and system constraints, the model adeptly handles multidimensional data, employing intelligent optimization algorithms to enhance its precision and robustness. It uniquely considers the interdependencies among parameters, such as complementarity and substitutability effects, ensuring a more accurate representation of real-world dynamics. To validate its efficacy, sensitivity analysis is conducted, showcasing the model's stability and adaptability across various scenarios. The findings underscore the potential of combining MILP and GA in addressing high-dimensional complex systems, providing a solid scientific foundation for future applications and refinements in optimization modeling techniques. This research not only advances the theoretical understanding but also offers practical insights into optimizing resource management in diverse fields.
genetic algorithms (GAs) are meta-heuristic algorithms that are used for solving constrained and unconstrained optimization problems, mimicking the process of natural selection in biological evolution. Due to the fact...
详细信息
genetic algorithms (GAs) are meta-heuristic algorithms that are used for solving constrained and unconstrained optimization problems, mimicking the process of natural selection in biological evolution. Due to the fact that GAs do not require the optimization function to be differentiable, they are suitable for application in cases where the derivative of the objective function is either unavailable or impractical to obtain numerically. This paper proposes a general purpose genetic algorithm toolkit, implemented in Python3 programming language, having only minimum dependencies in NumPy and Joblib, that handle some of the numerical and parallel execution details.
Time and energy efficiency is a highly relevant objective in high-performance computing systems, with high costs for executing the tasks. Among these tasks, evolutionary algorithms are of consideration due to their in...
详细信息
Time and energy efficiency is a highly relevant objective in high-performance computing systems, with high costs for executing the tasks. Among these tasks, evolutionary algorithms are of consideration due to their inherent parallel scalability and usually costly fitness evaluation functions. In this respect, several scheduling strategies for workload balancing in heterogeneous systems have been proposed in the literature, with runtime and energy consumption reduction as their goals. Our hypothesis is that a dynamic workload distribution can befitted with greater precision using metaheuristics, such as genetic algorithms, instead of linear regression. Therefore, this paper proposes anew mathematical model to predict the energy-time behaviour of applications based on multi-population genetic algorithms, which dynamically distributes the evaluation of individuals among the CPU-GPU devices of heterogeneous clusters. An accurate predictor would save time and energy by selecting the best resource set before running such applications. The estimation of the workload distributed to each device has been carried out by simulation, while the model parameters have been fitted in a two-phase run using another genetic algorithm and the experimental energy-time values of the target application as input. When the new model is analysed and compared with another based on linear regression, the one proposed in this work significantly improves the baseline approach, showing normalised prediction errors of 0.081 for runtime and 0.091 for energy consumption, compared to 0.213 and 0.256 shown in the baseline approach.
Supervised machine learning is widely researched nowadays. Several works have already been developed using genetic algorithms (GAs) for classification tasks evolving IF-THEN classification rules. Oftentimes, these met...
详细信息
Supervised machine learning is widely researched nowadays. Several works have already been developed using genetic algorithms (GAs) for classification tasks evolving IF-THEN classification rules. Oftentimes, these methods are built using integers and real values from one's chromosome structure. In this paper, new and important improvements are proposed to Non-linear Computation Evolutionary Environment (NLCEE), a GA-based rule-set generator proposed by Amaral and Hruschka. The proposed GA, called BIN-NLCEE, uses binary representation in its chromosome structure to simplify its mutation and also produce a higher search space. The main goal is to have a rule-set generator that produces simple and interpretable classification rules with good accuracy values and better converge rates. The BIN-NLCEE performance was compared with other GAs-based and four traditional classifiers in five medical domain datasets. The results showed a better convergence rate and higher fitness values for BIN-NLCEE when compared with the GA-based CEE and NLCEE. In 20 comparisons, BIN-NLCEE achieved better results in 9 (45%), and, according to the confidence interval, equivalent results were obtained in 11 (55%). In this way, BIN-NLCEE was better or equal to NLCEE and CEE in 100% of the comparisons. Also, BIN-NLCEE outperformed all traditional classifiers' results, i.e., achieved better results in 100% of comparisons.
Complex sociotechnical systems with multiple competing objectives and nonlinear dynamics pose significant challenges for policy optimization. Traditional simulation-based methods often struggle with high-dimensional p...
详细信息
Complex sociotechnical systems with multiple competing objectives and nonlinear dynamics pose significant challenges for policy optimization. Traditional simulation-based methods often struggle with high-dimensional policy spaces. This study addresses these challenges by combining genetic algorithms (GAs) with system dynamics (SD) modeling to optimize policy configurations in complex environments. Our approach merges SD's capacity to simulate intricate system behaviors with GA's prowess in multi-objective optimization. The SD model uses predefined decision variables to simulate system behavior, while GA iteratively adjusts these variables to find optimal policy configurations. We apply this hybrid approach to a media industry case study, focusing on balancing profit, competitiveness, and audience satisfaction. The research methodology integrates SD's ability to capture complex system behaviors with GA's strength in optimizing multiple objectives. An SD model simulates system behavior based on predefined decision variables representing key policy levers. GA then iteratively adjusts these variables based on fitness objectives derived from the SD model. This process evaluates performance and identifies the globally optimal policy configurations. The results show that the hybrid SD-GA framework significantly improves policy solutions compared to conventional methods. Sensitivity analysis confirms the optimized policies' robustness and comparative assessments highlight our approach's advantages in navigating complex policy spaces. This study introduces a new SD-GA framework that improves data-driven policy formulation in complex systems. It combines policy informatics and evolutionary algorithms to offer a comprehensive approach to multi-criteria decision-making. Future research could include long-term validation of this approach in the broadcasting industry to further refine and apply this methodology across various domains.
This paper presents a comprehensive framework for offline optimization of tuning parameters in unmanned aerial vehicle (UAV) flight controllers. The framework uses system identification to create a simplified flight d...
详细信息
This paper presents a comprehensive framework for offline optimization of tuning parameters in unmanned aerial vehicle (UAV) flight controllers. The framework uses system identification to create a simplified flight dynamics model, followed by control law matching to ensure the simulated controller's output closely replicates real-world autopilot commands. The optimization phase employs genetic algorithms to tune parameters based on a defined cost function that incorporates performance requirements. Each stage, from flight dynamics model development to optimization, is validated to ensure enhanced controller performance. Finally, real-world flight tests confirm the effectiveness of the optimized controller, demonstrating the validity of the proposed framework for autopilot tuning optimization.
Iris recognition is a critical component in biometric identification systems, known for its high accuracy and reliability. However, traditional methods often struggle with challenges related to feature extraction and ...
详细信息
Iris recognition is a critical component in biometric identification systems, known for its high accuracy and reliability. However, traditional methods often struggle with challenges related to feature extraction and classification, especially under varying conditions like lighting and occlusions. This paper addresses these challenges by proposing an enhanced iris recognition approach that combines Histogram Cut Selection (HCS) with genetic algorithms (GA). The HCS technique is employed for initial feature extraction and segmentation, which effectively isolates the most significant iris features while reducing noise and irrelevant data. Following this, genetic algorithms are applied to optimize the classification process by iteratively refining the decision boundaries, ensuring a robust and accurate recognition system. To further enhance segmentation accuracy, we introduce a recursive entropy discretization model. This model works in tandem with HCS to segment the iris with higher precision, leading to improved feature representation. The proposed method was tested on several benchmark iris datasets, demonstrating superior performance compared to traditional methods. Specifically, the recognition accuracy improved by 8.5%, and the computational efficiency increased by 15%, making this method highly suitable for real-time biometric identification applications. The integration of HCS with GA not only enhances the robustness of the system but also ensures its adaptability across different environments.
暂无评论