检索结果-内蒙古大学图书馆

A New Prediction Strategy for Dynamic Multi-objective optimization Using Diffusion Model

IEEE Transactions on Evolutionary Computation 2025年

作者： Wang, Feng Xie, Jinsong Zhou, Aimin Tang, Ke Wuhan University School of Computer Science Hubei430072 China East China Normal University Shanghai Key Laboratory of Multidimensional Information Processing The Department of Computer Science and Technology Shanghai200241 China Southern University of Science and Technology Department of Computer Science and Engineering Shenzhen518055 China

To solve dynamic multi-objective optimization problems (DMOPs), the optimization algorithms are required to track the movement of the Pareto set after the environmental changes effectively. Many prediction-based dynamic multi-objective evolutionary algorithms (DMOEAs) have been proposed to address this challenge by utilizing environmental information for population reinitialization. However, when environmental changes are complex, irregular, and severe, the solutions and information during the evolution process often contain noise, making it difficult for prediction-based DMOEAs to accurately predict and reinitialize the population. To address this issue, we propose a novel dynamic multi-objective evolutionary algorithm (DM-DMOEA) which uses a diffusion model based prediction strategy. In DM-DMOEA, to improve the prediction accuracy, the diffusion model is introduced to extract the relationships of high-quality solutions and reinitialize the population, and a PS estimation method is employed to integrate both historical and new environmental information, providing a set of high-quality solutions for diffusion model training. To speed up the response time, a variational autoencoder (VAE) is used to map the decision space to a latent space, which can reduce the diffusion model size and accelerate the diffusion process. To evaluate the effectiveness of the proposed DM-DMOEA on DMOPs, comprehensive experiments are conducted on several benchmarks and a practical problem. The results show that the DM-DMOEA outperforms other four state-of-the-art DMOEAs in most cases. © 1997-2012 IEEE.

关键词： optimization algorithms

来源：评论

学校读者我要写书评

暂无评论

Joint Request Offloading and Resource Allocation for Long-Term Utility optimization in Collaborative Edge Inference with Time-Coupled Resources

引用

IEEE Transactions on Network Science and Engineering 2025年

作者： Huang, Jiale Wu, Jigang Wu, Yalan Wu, Jiaxin The School of Computer Science and Technology Guangdong University of Technology Guangzhou510006 China The School of Integrated Circuits Guangdong University of Technology Guangzhou510006 China

Extensive research on edge inference has devoted in optimizing service performance for users. However, recent studies have overlooked the desired utility of application service provider (ASP), which is crucial for achieving long-term service provisioning. Besides, efficient request offloading and resource allocation are essential for optimizing long-term utility of ASP in dynamic networks with time-coupled resources. To address these issues, this paper formulates a long-term utility optimization problem in collaborative edge inference system. The objective is to maximize the long-term average utility of ASP, by jointly optimizing request offloading and resource allocation, under the time-coupled resource constraints. To solve the problem, a Lyapunov based online algorithm is proposed to decompose it into a series of one-slot deterministic problems by decoupling the time-coupled resource constraints. Only the current network states are required for one-slot problem. Then, the one-slot problem is converted into a master request offloading problem with an inner resource allocation problem. A distributed algorithm is proposed to derive the optimal decision to inner problem, while a coalition based algorithm is proposed to seek the stable solution to master problem. Experimental results show that, the proposed algorithm outperforms baseline algorithms for most cases, in terms of long-term average utility of ASP. © 2013 IEEE. All rights reserved.

关键词： optimization algorithms

来源：评论

学校读者我要写书评

暂无评论

A niching differential evolution with Hilbert curve for multimodal multi-objective optimization

引用

Swarm and Evolutionary Computation 2025年 95卷

作者： Li, Guosen Li, Wenfeng He, Lijun Gao, Cong School of Transportation and Logistics Engineering Wuhan University of Technology Wuhan430063 China Sanya Science and Education Innovation Park Wuhan University of Technology Hainan572000 China State Key Laboratory of Maritime Technology and Safety Wuhan University of Technology Wuhan430063 China

Multimodal multi-objective optimization problems have a many-to-one relationship between the decision space and the objective space. That is, distinct solutions in the decision space share the same objective value. How to coordinate population convergence and diversity while locating multimodal solutions is a challenging research topic. Some evolutionary algorithms using niching techniques have been reported in the literature. These algorithms prefer to induce multiple niches based on population information. Owing to the impact of convergence-first principle, the population tends to gather in easier-to-search regions, making it tough to yield more dispersed solutions in different niches. To remedy this situation, this paper proposes a niching differential evolution with Hilbert curve. First, a neighborhood-driven reproduction method is presented based on Hilbert curve, which features a two-layer architecture to capture promising regions and identify multimodal solutions. Second, a convergence-based density indicator is designed as a selection criterion to distinguish between convergence solutions and diversity solutions in the decision space. Moreover, fifteen intricate multimodal multi-objective test functions are devised. The experiments are performed on a series of test functions and a map-based practical problem. Empirical results attest that the proposed algorithm is competitive in dealing with multimodality compared with ten multimodal multi-objective algorithms. © 2025

关键词： optimization algorithms

来源：评论

学校读者我要写书评

暂无评论

FROM: A Fish Recognition-Inspired optimization Method for Multi-Agent Decision-Making Problems with a Fluid Environment

引用

Biomimetics 2025年第4期10卷 215-215页

作者： Wang, Yuchen Sun, Lei School of Software North China University of Water Resources and Electric Power Zhengzhou450046 China

Underwater multi-agent systems face critical hydrodynamic constraints that significantly degrade the performance of conventional constraint optimization algorithms in dynamic fluid environments. To meet the needs of underwater multi-agent applications, a fish recognition-inspired optimization method (FROM) is proposed in this paper. The proposed method introduces the characteristics of fish recognition. There are two major improvements in the proposed method: the neighbor topology improvement based on vision recognition and the learning strategies improvement based on hydrodynamic recognition. The computational complexity of the proposed algorithm was analyzed, and it was found to be acceptable. The statistical analysis of the experimental results shows that the FROM algorithm performs better than other algorithms in terms of minimum, maximum, standard deviation, mean, and median values calculated from objective functions. With solid experiment results, we conclude that the proposed FROM algorithm is a better solution to solve multi-agent decision-making problems with fluid environment constraints. © 2025 by the authors.

关键词： optimization algorithms

来源：评论

学校读者我要写书评

暂无评论

Integration of adaptive projection BFGS and inertial extrapolation step for nonconvex optimization problems and its application in machine learning

引用

Journal of the Franklin Institute 2025年第7期362卷

作者： Yuan, Gonglin Yang, Yuehan Li, Yong Zhao, Xiong Meng, Zehong Guangxi University Nanning Guangxi530004 China Department of Mathematics Baise University Baise Guangxi533000 China School of Mathematics and Statistics Zhejiang University of Finance and Economics Hangzhou Zhejiang310018 China

With the rapid development of machine learning and large data technologies, large-scale optimization problems become more and more common, and traditional optimization algorithms face the challenges of computational complexity and memory consumption. In this paper, we combine the adaptive projection BFGS method with the inertial extrapolation technique, and propose a BFGS algorithm combining inertial extrapolation and adaptive projection, which enhances the search capability of the algorithm, ensures that there is a suitable descent direction in each iteration, and reduces the oscillation in the iteration. Under certain mild conditions and weak Wolfe–Powell (WWP) conditions, the algorithm demonstrates both global convergence and superlinear convergence rates for nonconvex unconstrained optimization problems. In addition, to solve large-scale optimization problems, we propose an inertial extrapolation adaptive projection BFGS algorithm based on variance reduction, which performs well in dealing with large-scale datasets and offers new insights to address the limitations of traditional algorithms. Finally, we evaluate their performance and efficiency using classical numerical experiments and applications in machine learning models. © 2025 The Franklin Institute

关键词： optimization algorithms

来源：评论

学校读者我要写书评

暂无评论

A Multi-Fidelity Genetic Algorithm for Hyperparameter optimization of Deep Neural Networks

引用

IEEE Transactions on Evolutionary Computation 2025年

作者： Moya, Antonio R. Ventura, Sebastian Córdoba Spain

Hyperparameter optimization on Machine Learning models is crucial for their correct refinement. For complex big models (such as Deep Learning models), in which a single training model is supposed to have a very high computational cost, this optimization sometimes becomes unfeasible. Multi-fidelity optimization algorithms are a solution to alleviate this computational cost of optimizing hyperparameters of Deep Learning models. In this scope, we propose GAMF2O, a new multi-fidelity algorithm that relies on a genetic algorithm. This paper clearly defines how to adapt the evolutionary scenario to follow the multi-fidelity approach, and we propose a new scheme to evaluate each individual based on the use of two objectives: the result of the low-fidelity evaluation and the learning capacity, with the use of the latter being novel during the evaluation process. Our experimental section allows us to show how our proposal improves the state-of-the-art in different classification and regression problems. © 1997-2012 IEEE.

关键词： optimization algorithms

来源：评论

学校读者我要写书评

暂无评论

Enclosing 95% confidence area and volume to center of pressure and center of mass in posturography using optimization algorithm and coordinate ascent methods

引用

Computers in Biology and Medicine 2025年 185卷 109563-109563页

作者： Alighanbari, Mohsen Alighanbari, Sina Griffin, Lisa Department of Kinesiology and Health Education University of Texas AustinTX78712 United States Department of Systems Design Engineering University of Waterloo WaterlooONN2L3G1 Canada Autonomy Department Robotic Assistance Devices Inc. WaterlooONN2V1K8 Canada

Balance control has been evaluated using center of pressure (CoP) and center of mass (CoM). One of the most common approaches in stabilometry is enclosing ellipse to 95% of data using principal component analysis (PCA) or covariance methods. However, these methods have limitations, including normality assumption, lack of accuracy, and sample size influence. This study aims to use a coordinate-ascent optimization algorithm to address the above limitations. Fifteen healthy young adults were recruited and performed six sit-to-stand trials. The optimization algorithm, data trimmer, PCA, and covariance methods were used. For robust analysis, 15000 CoM and CoP data were simulated for five distribution types. Repeated measure ANOVA was used to evaluate the difference in area and volume enclosed using different methods. The quality of fit (accuracy) and the precision of the methods were assessed using the percentage goodness of fit and coefficient of variation. Robustness was also explored with the 95% confidence interval of the mean absolute error and coefficient variation of error. The optimization enclosed smaller ellipse/ellipsoids to CoP and CoM data than other methods with higher accuracy, precision, and robustness. Although the optimization enclosed larger ellipses than the data trimmer, its results were more accurate and precise. The coefficient of variation of error indicated that the optimization performs with marginally higher estimation error for leptokurtic data compared to the data trimmer. Overall, our results demonstrated that the optimization method performs better than other methods, leading to higher sensitivity in detecting changes in CoP and CoM due to pathological conditions. © 2024 Elsevier Ltd

关键词： optimization algorithms

来源：评论

学校读者我要写书评

暂无评论

An optimization method for corn planting parameters based on mutation breeding sea horse optimization algorithm

引用

Computers and Electronics in Agriculture 2025年 236卷

作者： Bei, Jinling Wang, Jiquan Zhang, Hongyu College of Engineering Northeast Agricultural University Harbin150030 China Rongtong Human Resources Development Co. Ltd Beijing102600 China

In response to the black box problem of optimizing corn planting parameters, and considering the shortcomings of traditional methods such as low fitting accuracy and susceptibility to local optima, a MBSHO-BPNN method based on mutation breeding seahorse optimization algorithm (MBSHO) and BP neural network (BPNN) is proposed. Firstly, the standard sea horse optimization algorithm (SHO) and its improved MBSHO are introduced, including control parameter improvement, spiral movement update, and mutation breeding mechanism. Subsequently, a series of extensive numerical experiments are conducted to systematically evaluate the effectiveness of MBSHO components and related parameters. MBSHO is also compared to other algorithms using the CEC 2017 tests on problems of different sizes. The findings indicated that MBSHO demonstrated superior performance. In the effective verification of MBSHO-BPNN, this method outperforms other comparative methods to fit accuracy and optimization results for unconstrained and linearly constrained optimization problems. Ultimately, the MBSHO − BPNN was applied to the optimization of corn planting parameters, and the optimal parameter combination was obtained: the planting density is 9.23 × 104/hm2, nitrogen fertilizer application rate is 138.72 kg/hm2,phosphorus fertilizer application rate is 86.53 kg/hm2, and the potassium fertilizer application rate is 70.32 kg/hm2. Under this configuration, the corn yield reached 16,303.56 kg/hm2, which is significantly higher than that of other methods. The relative error of the actual average yield is only − 0.6757 %. This method not only provides an efficient solution to the agricultural black-box optimization problem but also exhibits potential for broader nonlinear optimization challenges. © 2025 Elsevier B.V.

关键词： optimization algorithms

来源：评论

学校读者我要写书评

暂无评论

Multi-task snake optimization algorithm for global optimization and planar kinematic arm control problem

引用

PeerJ Computer Science 2025年 11卷 e2688-e2688页

作者： Li, Qingrui Zhou, Yongquan Luo, Qifang College of Artificial Intelligence Guangxi Minzu University Guangxi Nanning China Guangxi Key Laboratories of Hybrid Computation and IC Design Analysis Guangxi Nanning China

Multi-task optimization (MTO) algorithms aim to simultaneously solve multiple optimization tasks. Addressing issues such as limited optimization precision and high computational costs in existing MTO algorithms, this article proposes a multitask snake optimization (MTSO) algorithm. The MTSO algorithm operates in two phases: first, independently handling each optimization problem;second, transferring knowledge. Knowledge transfer is determined by the probability of knowledge transfer and the selection probability of elite individuals. Based on this decision, the algorithm either transfers elite knowledge from other tasks or updates the current task through self-perturbation. Experimental results indicate that, compared to other advanced MTO algorithms, the proposed algorithm achieves the most accurate solutions on multitask benchmark functions, the five-task and 10-task planar kinematic arm control problems, the multitask robot gripper problem, and the multitask car side-impact design problem. The code and data for this article can be obtained from: https://***/10.5281/zenodo.14197420. Copyright 2025 Li et al. Distributed under Creative Commons CC-BY 4.0

关键词： optimization algorithms

来源：评论

学校读者我要写书评

暂无评论

Integrating life cycle assessment in multi-objective optimization of green hydrogen systems: a review of literature and methodological challenges

引用

RENEWABLE & SUSTAINABLE ENERGY REVIEWS 2025年 217卷

作者： Chavez, Diego Larrahondo Azzaro-Pantel, Catherine Montignac, Florent Ruby, Alain Univ Grenoble Alpes CEA Liten DTCH F-38000 Grenoble France Univ Toulouse CNRS Lab Genie Chim INPTUPS Toulouse France

As global demand for sustainable energy solutions intensifies, reducing the environmental impact of energy systems while maintaining cost-effectiveness has become imperative, particularly for emerging energy carriers like hydrogen. This article presents a comprehensive review of life cycle assessment (LCA) and multi-objective optimization (MOO) in the context of green hydrogen (H2) systems, focusing on key contributions from 2019 to 2023. The review categorizes the literature into three areas: (i) Recent LCA studies on green H2 production, (ii) MOO studies that include LCA of renewable energy systems, (iii) Research integrating environmental criteria into MOO for green H2 systems. The review highlights significant gaps, particularly the limited depth of environmental detail in LCAs integrated into MOO, compared to standalone environmental LCAs. Additionally, the analysis reveals a lack of end-of-life considerations across all study categories. In terms of optimization, most research adopts a bi-objective framework, typically focusing on economic performance and a single environmental metric (often CO2 emissions), thereby constraining the development of more comprehensive, sustainable systems. The article concludes by offering recommendations: future LCAs of hydrogen systems should incorporate advanced optimization algorithms, broaden their scope to cradle-to-grave assessments, and utilize prospective methods for better anticipation of future environmental impacts.

关键词： Green hydrogen Life cycle assessment (LCA) Multi-objective optimization Environmental impacts optimization algorithms Renewable energy

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：