Large Language Models (LLMs) have demonstrated remarkable advancements across diverse domains, manifesting considerable capabilities in evolutionary computation, notably in generating new solutions and automating algo...
详细信息
Large Language Models (LLMs) have demonstrated remarkable advancements across diverse domains, manifesting considerable capabilities in evolutionary computation, notably in generating new solutions and automating algorithm design. Surrogate-assisted selection plays a pivotal role in evolutionary algorithms (EAs), especially in addressing expensive optimization problems by reducing the number of real function evaluations. However, whether LLMs can serve as surrogate models remains an unknown. In this study, we propose a novel surrogate model based purely on LLM inference capabilities, eliminating the need for training. Specifically, we formulate model-assisted selection as a classification problem or a regression problem, utilizing LLMs to directly evaluate the quality of new solutions based on historical data. This involves predicting whether a solution is good or bad, or approximating its value. This approach is then integrated into EAs, termed LLM-assisted EA (LAEA). Detailed experiments compared the visualization results of 2D data from 9 mainstream LLMs, as well as their performance on 5-10 dimensional problems. The experimental results demonstrate that LLMs have significant potential as surrogate models in evolutionary computation, achieving performance comparable to traditional surrogate models only using inference. This work offers new insights into the application of LLMs in evolutionary computation. Code is available at: https://***/hhyqhh/***.
This paper develops a dynamic optimization methodology based on direct numerical methods, for the bioethanol fed-batch production from glucose and fructose as a substrate. The mathematical model that governs the proce...
详细信息
This paper develops a dynamic optimization methodology based on direct numerical methods, for the bioethanol fed-batch production from glucose and fructose as a substrate. The mathematical model that governs the process consists of six differential equations and is highly nonlinear. The proposed strategy uses the Fourier trigonometric basis and normalized orthogonal polynomials for substrate feeding rate parameterization. Then, evolutionary algorithms and gradient methods are combined to search parameters that generate the best control action. This parameterization methodology requires a minimum number of parameters to optimize. Also, the continuous and differentiable nature of the optimal profile enables its direct implementation in the physical process, eliminating the necessity for filtering or smoothing it. In addition, they are ideal for bioprocesses, in which it is preferable to avoid abrupt changes in the operating modes of the process to promote cell growth. As a result, using only 3 parameters, a 3.5% increase in ethanol production was achieved, while the reference uses at least 10 parameters and provides a stepped feed profile. The simulations have yielded promising results, making this proposal an alternative with excellent potential for process optimization.
The interest in understanding nanofluid density's impact on heat transfer and fluid flow behaviors has driven the need for accurate density values. Artificial intelligence techniques for predicting nanofluid densi...
详细信息
The interest in understanding nanofluid density's impact on heat transfer and fluid flow behaviors has driven the need for accurate density values. Artificial intelligence techniques for predicting nanofluid density provide a cost-effective and efficient alternative to labor-intensive lab experiments. In the current research, four distinct models based on the Radial Basis Function (RBF) Neural Network were developed and implemented on an extensive and comprehensive databank comprising 4004 experimental data-points gathered from multiple available sources. Ant Colony Optimization (ACO), Particle Swarm Optimization (PSO), Imperialist Competitive Algorithm (ICA), and Genetic Algorithm (GA) were used separately to optimize the neural network. The provided databank introduces 95 varieties of mono-nanofluids, including 16 types of nano-particles and 11 types of base- fluids. The target/dependent variable in this research is the density of mono-nanofluids (rho nf), whereas the input/ independent variables include the average nano-particle diameter (dnp), nano-particle mass concentration (phi m), temperature (T), pressure (P), nano-particle density (rho np), and base-fluid density (rho bf). Various analyses of the four models confirmed the RBF-ACO model's robustness and superiority. Key statistical indicators for this model indicated an Average Absolute Percent Relative Error (AAPRE) of 0.5391%, a Standard Deviation (SD) of 0.0099, and a Coefficient of Determination (R2) of 0.9818. Sensitivity analysis for the superior model identified key input-output variables with high relevancy factors (r-values). Notably, variables phi m and rho bf had maximum rvalues close to 0.70, indicating their significant role in predicting mono-nanofluids' density. In addition, a leverage statistical approach was utilized to determine possible outliers and the applicability domain of the RBFACO model.
evolutionary algorithm- based size optimization of plane and space truss structures is studied for sequential loading scenarios. Size optimization of a truss is a search for the most suitable cross-sectional areas of ...
详细信息
evolutionary algorithm- based size optimization of plane and space truss structures is studied for sequential loading scenarios. Size optimization of a truss is a search for the most suitable cross-sectional areas of the truss members from the available design spaces. The novelty of the present work lies in developing an evolution- based algorithm that can consider the sequential loading scenario while designing the planner and space trusses. Fortran- based computer programming has been developed for the proposed optimization method considered here. A standard displacement- based finite element method is implemented to obtain the trusses' nodal displacement and elemental stresses. To assess the performance of the proposed algorithm three plane trusses (3- bar, 18- bar, and 200- bar) and three space trusses (22- bar, 25- bar, and 72- bar) with various displacement and stress constraints under sequential loading have been considered. The optimum weights obtained from all the problems considered here have been analyzed and compared with the same, as obtained from other methods mentioned in the existing literature. It is important to note here that the results are in close agreement and in some cases optimum weights obtained from the present study are even better than the earlier results. Finally, a real- life design problem of an industrial roof truss has been considered to assess the applicability of the proposed algorithm. The convergence study and optimum cross- section of the truss have been reported as a case study.
Doubly-stochastic matrices play a vital role in modern applications of complex networks such as tracking and decentralized state estimation, coordination and control of autonomous agents. A central theme in all of the...
详细信息
Doubly-stochastic matrices play a vital role in modern applications of complex networks such as tracking and decentralized state estimation, coordination and control of autonomous agents. A central theme in all of the above is consensus, that is, nodes reaching agreement about the value of an underlying variable (e.g. the state of the environment). Despite the fact that complex networks have been studied thoroughly, the communication graphs are usually described by symmetric matrices due to their advantageous theoretical properties. We do not yet have methods for optimizing generic doubly-stochastic matrices. In this paper, we propose a novel formulation and framework, EvoDSM, for achieving fast linear distributed averaging by: (a) optimizing the weights of a fixed graph topology, and (b) optimizing for the topology itself. We are concerned with graphs that can be described by positive doubly-stochastic matrices. Our method relies on swarm and evolutionary optimization algorithms and our experimental results and analysis showcase that our method (1) achieves comparable performance with traditional methods for symmetric graphs, (2) is applicable to non-symmetric network structures and edge weights, and (3) is scalable and can operate effectively with moderately large graphs without engineering overhead.
Smart technology have end up an increasing number of vital in today's rapidly evolving generation panorama. Automation, records-pushed decision-making, and streamlined operations are all being revolutionized by me...
详细信息
Benchmark problems have been fundamental in advancing our understanding of the dynamics and design of multi-objective evolutionary optimization algorithms. Within the binary domain, there is a lack of multi-objective ...
详细信息
Benchmark problems have been fundamental in advancing our understanding of the dynamics and design of multi-objective evolutionary optimization algorithms. Within the binary domain, there is a lack of multi-objective benchmark problems that can help further research on constrained optimization. This paper presents highly configurable benchmark problems for constrained binary multi-objective optimization combining SAT Constraints, constructed from satisfiability clauses, and MNK-Landscapes. The benchmark problems are scalable in the number of equality and inequality constraints, feasibility-hardness, number of objectives, number of variables, and epistasis between variables. This paper studies how SAT Constraints affect the distribution of feasible solutions in objective and decision spaces and illustrates their impact on the performance and dynamics of multi-objective evolutionary algorithms when solving SAT Constrained MNK-Landscapes.
This paper studies the use of machine learning models for multiobjective optimization of vaccinations used to control an epidemic spreading in a graph representing contacts between individuals. Graph nodes are pa...
详细信息
The assessment of apple quality is pivotal in agricultural production management, and apple ripeness is a key determinant of apple quality. This paper proposes an approach for assessing apple ripeness from both struct...
详细信息
The assessment of apple quality is pivotal in agricultural production management, and apple ripeness is a key determinant of apple quality. This paper proposes an approach for assessing apple ripeness from both structured and unstructured observation data, i.e., text and images. For structured text data, support vector regression (SVR) models optimized using the Whale Optimization Algorithm (WOA), Grey Wolf Optimizer (GWO), and Sparrow Search Algorithm (SSA) were utilized to predict apple ripeness, with the WOA-optimized SVR demonstrating exceptional generalization capabilities. For unstructured image data, an Enhanced-YOLOv8+, a modified YOLOv8 architecture integrating Detect Efficient Head (DEH) and Efficient Channel Attention (ECA) mechanism, was employed for precise apple localization and ripeness identification. The synergistic application of these methods resulted in a significant improvement in prediction accuracy. These approaches provide a robust framework for apple quality assessment and deepen the understanding of the relationship between apple maturity and observed indicators, facilitating more informed decision-making in postharvest management.
Quantum error correction and the use of quantum error correction codes are likely to be essential for the realization of practical quantum computing. Because the error models of quantum devices vary widely, quantum co...
详细信息
Quantum error correction and the use of quantum error correction codes are likely to be essential for the realization of practical quantum computing. Because the error models of quantum devices vary widely, quantum codes that are tailored for a particular error model may have much better performance. In this work, we present a novel evolutionary algorithm that searches for an optimal stabilizer code for a given error model, number of physical qubits, and number of encoded qubits. We demonstrate an efficient representation of stabilizer codes as binary strings, which allows for random generation of valid stabilizer codes as well as mutation and crossing of codes. Our algorithm finds stabilizer codes whose distance closely matches the best-known-distance codes of Grassl (2007) for n <= 20 physical qubits. We perform a search for optimal distance Calderbank-Steane-Shor codes and compare their distance to the best known codes. Finally, we show that the algorithm can be used to optimize stabilizer codes for biased error models, demonstrating a significant improvement in the undetectable error rate for [[12,1]](2) codes versus the best-known-distance code with the same parameters. As part of this work, we also introduce an evolutionary algorithm QDistEvol for finding the distance of quantum error correction codes.
暂无评论