检索结果-内蒙古大学图书馆

4th International Conference on Artificial Intelligence, Automation and High Performance Computing, AIAHPC 2024

作者： Wang, Yijun Chen, Jiaqi Zhuang, Dawei School of Electronic Information Engineering Changchun University of Science and Technology China

ISBN: (纸本)9798400710049

In order to improve the coverage of UAV AD hoc network, reduce the loss of UAV in operation and the number of UAV involved, a network coverage optimization algorithm based on improved moth extinguishing algorithm is proposed. Firstly, the artificial potential energy method is applied to avoid obstacles before the global search, secondly, the nonlinear dynamic adaptive step factor is introduced to expand the moth search range at the beginning of the search stage, and finally, the dynamic adjustment strategy of flight direction is integrated to improve the global optimization ability during the search process, and then the optimal solution is modified by Cauchy. The simulation results show that the convergence and stability of the proposed algorithm are superior to the other four algorithms, and the connectivity between UAV nodes is improved, and the coverage rate is 18.32%, 5.48%, 5.09% and 10.6% higher than that of the other four algorithms, respectively, which effectively reduces the loss and the number of UAVs used and reduces the cost. © 2024 Copyright held by the owner/author(s).

关键词： optimization algorithms

来源：评论

学校读者我要写书评

暂无评论

Improving Proximal Policy optimization with Hellinger Divergence 2

Improving Proximal Policy Optimization with Hellinger Diverg...

引用

2nd International Conference on Algorithm, Image Processing and Machine Vision, AIPMV 2024

作者： Zhou, Jiayi Feng, Jixing Chang, Zhiming Shandong University at Weihai School of Mathematics and Statistics Shandong Weihai China

ISBN: (纸本)9798350390254

Proximal policy optimization is a reinforcement learning algorithm widely used for training policies. Its goal is to learn how to make decisions through interaction with the environment to maximize some kind of cumulative reward. This article proposes an improved Proximal Policy optimization (PPO) algorithm based on Hellinger distance, aiming to enhance the stability and effect of policy optimization in reinforcement learning. By introducing the measure of Hellinger distance when updating the policy, we limit the changes in the policy distribution in each update, thereby improving the convergence of the algorithm and the stability of the training process. Experiments have shown that this algorithm performs well in handling tasks in the action space, and achieves better performance and convergence speed than the traditional PPO algorithm. This research has certain theoretical and practical value for improving and expanding reinforcement learning algorithms based on probability distribution. © 2024 IEEE.

关键词： optimization algorithms

来源：评论

学校读者我要写书评

暂无评论

Decomposition of Difficulties in Complex optimization Problems using a Bilevel Approach 13

Decomposition of Difficulties in Complex Optimization Proble...

引用

13th IEEE Congress on Evolutionary Computation, CEC 2024

作者： Sinha, Ankur Pujara, Dhaval Singh, Hemant Kumar Centre for Data Science and Ai Indian Institute of Management Gujarat Ahmedabad380015 India School of Engineering and Technology The University of New South Wales CanberraACT2610 Australia

ISBN: (纸本)9798350308365

Practical optimization problems may contain differ-ent kinds of difficulties that are often not tractable if one relies on a particular optimization method. Different optimization approaches offer different strengths that are good at tackling one or more difficulty in an optimization problem. For instance, evolutionary algorithms have a niche in handling complexities like discontinuity, non-differentiability, discreteness and non-convexity. However, evolutionary algorithms may get computationally expensive for mathematically well behaved problems with large number of variables for which classical mathematical programming approaches are better suited. In this paper, we demonstrate a decomposition strategy that allows us to synergistically apply two complementary approaches at the same time on a complex optimization problem. Evolutionary algorithms are useful in this context as their flexibility makes pairing with other solution approaches easy. The decomposition idea is a special case of bilevel optimization that separates the difficulties into two levels and assigns different approaches at each level that is better equipped at handling them. We demonstrate the benefits of the proposed decomposition idea on a wide range of test problems. © 2024 IEEE.

关键词： optimization algorithms

来源：评论

学校读者我要写书评

暂无评论

Enabling Dual Subpopulations and Clustering for Multimodal Multiobjective optimization 13

Enabling Dual Subpopulations and Clustering for Multimodal M...

引用

13th IEEE Congress on Evolutionary Computation, CEC 2024

作者： Su, Yu-Cheng Chen, Yi-Ruei Ting, Chuan-Kang National Tsing Hua University Department of Power Mechanical Engineering Hsinchu30013 Taiwan National Tsing Hua University Department of Computer Science Hsinchu30013 Taiwan

ISBN: (纸本)9798350308365

Multimodal multiobjective optimization problems (MMOPs) present a specific category of multiobjective problems. Unlike conventional multiobjective problems where Pareto opti-mal solutions gather in one or few regions in the decision space, MMOPs aim to find solutions spread across various regions. In particular, these solutions may share similar locations in objective space, making the problem more complex and challenging. Most multiobjective optimization evolutionary algorithms (MOEAs) barely preserve diversity in decision space, whereas multimodal multiobjective optimization evolutionary algorithms (MMOEAs) require additional techniques such as clustering to maintain diversity in the decision space. In this study, we propose a new method that enables two sub populations and a clustering technique for diversity preservation. Furthermore, different clustering methods are compared to investigate their impact on performance. The experimental results show that the proposed method achieves better inverted generational distance in decision space (IGDX), indicating the effectiveness of the diversity preservation approach. In addition, the varying IGDX and inverted generational distance (IGD) obtained from different clustering methods reveal their different strengths in solving MMOPs. © 2024 IEEE.

关键词： optimization algorithms

来源：评论

学校读者我要写书评

暂无评论

optimization of Three-Stage Heat Exchanger Design Using Modified Differential Evolution Algorithm 1

引用

2nd Biennial International Symposium on Fluids and Thermal Engineering, FLUTE 2023

作者： Gawai, I.R. Lalwani, D.I. SVNIT Gujarat Surat395007 India

ISBN: (数字)9789819724819

ISBN: (纸本)9789819724802

The size plays a decisive part in the selection of a heat exchanger. A heat exchanger with a smaller size with equal or higher thermal performance is always desirable. In this paper, the amended differential evolution algorithm (ADEA) is modified (modified-ADEA), and utilized to optimize a three-stage heat exchanger (TSHE) design problem. Also, a novel penalty assignment method is introduced which assigns a penalty based on the degree of constraint violation and total constraints violated by the individual. The three-stage heat exchanger is optimized to obtain the same performance with minimum heat transfer area. The optimized solution is compared with the optimum solution reported by algorithms ADEA, BA, HS, and GA. The modified ADEA found the best optimum results among the compared algorithms. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2024.

关键词： optimization algorithms

来源：评论

学校读者我要写书评

暂无评论

optimization of Motion Distribution for Modular Robots Based on Grey Wolf Algorithm

Optimization of Motion Distribution for Modular Robots Based...

引用

2024 IEEE International Conference on Robotics and Biomimetics, ROBIO 2024

作者： Yu, Longchang Chen, Gang Liang, Zhengyu Fei, Junting School of Automation Beijing University of Posts and Telecommunications Beijing China

ISBN: (纸本)9781665481090

Modular robots are limited in terms of working duration and performance due to the short-comings of existing motion planning algorithms in energy consumption allocation. This paper proposes an optimization strategy for the motion distribution of modular robots based on the Grey Wolf optimization (GWO) algorithm, aiming to improve the energy utilization and working duration of the robot. This paper emphasizes the energy consumption of each module as a critical metric, and addresses a high-dimensional discrete binary optimization problem based on the GWO algorithm to achieve balanced energy distribution, thereby significantly extending the working duration of the robot. The experimental part verifies the effectiveness of the proposed method for modular robots with different topological structures and different motion trajectories. The results show that the optimization strategy can significantly reduce the energy consumption difference between modules and increase the working duration. © 2024 IEEE.

关键词： optimization algorithms

来源：评论

学校读者我要写书评

暂无评论

Hybrid optimization Algorithm of Differential Evolution Algorithm and Artificial Bee Colony Algorithm 6

Hybrid Optimization Algorithm of Differential Evolution Algo...

引用

6th International Conference on Communications, Information System and Computer Engineering, CISCE 2024

作者： Song, Xiaoyu Zhu, Yanlin Zhao, Ming School of Computer Science and Engineering Shenyang Jianzhu University Shenyang China

ISBN: (纸本)9798350353174

In order to make full use of the advantages of artificial bee colony algorithm (ABC) and differential evolution algorithm (DE), a hybrid optimization algorithm based on differential evolution algorithm and artificial bee colony algorithm (HODEABC) is proposed in this paper. Firstly, we improve the strategy of the DE and ABC algorithm, utilizing two strategies based on the neighborhood information and the whole population information to adapt the strategy according to the improvement rate. Secondly, we employ a hybrid algorithm based on population division to evaluate performance based on the accumulated fitness value, and allocate more computing resources to algorithms demonstrating superior performance. Finally, compared with the improved algorithm on the CEC2014 test set and 22 standard function test sets, the results show that the performance of the proposed algorithm is comparable to these improved algorithms. © 2024 IEEE.

关键词： optimization algorithms

来源：评论

学校读者我要写书评

暂无评论

Novel Iteratively Preconditioned Gradient-Descent Algorithm via Successive Over-Relaxation Formulation

引用

IEEE CONTROL SYSTEMS LETTERS 2024年 8卷 3105-3110页

作者： Liu, Tianchen Chakrabarti, Kushal Chopra, Nikhil Univ Maryland Coll Pk Dept Mech Engn College Pk MD 20742 USA Tata Consultancy Serv Res Div Data & Decis Sci Mumbai 400607 India

We devise a novel quasi-Newton algorithm for solving unconstrained convex optimization problems. The proposed algorithm is built on our previous framework of the iteratively preconditioned gradient-descent (IPG) algorithm. IPG utilized Richardson iteration to update a preconditioner matrix that approximates the inverse of the Hessian matrix. In this letter, we substitute the Richardson iteration with a successive over-relaxation (SOR) formulation. The convergence guarantee of the proposed algorithm and its theoretical improvement over vanilla IPG are presented. The algorithm is used in a mobile robot position estimation problem for numerical validation using a moving horizon estimation (MHE) formulation. Compared with IPG, the results demonstrate an improved performance of the proposed algorithm in terms of computational time and the number of iterations needed for convergence, matching our theoretical results.

关键词： Estimation optimization algorithms robotics Estimation optimization algorithms robotics

来源：评论

学校读者我要写书评

暂无评论

Application of Multi-objective optimization Algorithm in Joint Operation of Sluice Groups

Application of Multi-objective Optimization Algorithm in Joi...

引用

2024 International Conference on Telecommunications and Power Electronics, TELEPE 2024

作者： Chen, Taiwu Pan, Zhifu Xu, Guanzhong Peng, Chen Liu, Erjun Lianyungang Linhong Water Conservancy Project Management Office Jiangsu Lianyungang222006 China Lianyungang Urban Water Engineering Management Office Jiangsu Lianyungang222005 China Guanyun County Dovetail Gate Management Guanyun Jiangsu222200 China

ISBN: (纸本)9798350369212

The purpose of this study is to construct a joint dispatching model of sluice groups by using multi-objective optimization algorithm, and its effectiveness is verified by simulation experiments. Methodologically, firstly, the premise and hypothesis of the model are clarified, and a mathematical model including water balance, water level and flow rate is constructed, and the corresponding constraints are set. Then, the multi-objective optimization algorithm is applied to find the operation strategy that can optimize the water resources utilization efficiency, flood control capacity and power generation benefit at the same time. The experimental results show that the model can find one or more optimized sluice operation strategies under different hydrological conditions and dispatching requirements, which significantly improves the overall utilization efficiency of water resources and achieves a good balance between flood control, power generation and shipping. Compared with other common sluice dispatching methods, this model performs better in comprehensive performance. This study provides a scientific basis for the design and operation of water conservancy projects, and has important theoretical value and practical significance. © 2024 IEEE.

关键词： optimization algorithms

来源：评论

学校读者我要写书评

暂无评论

DM-SARAH: A Variance Reduction optimization Algorithm for Machine Learning Systems

DM-SARAH: A Variance Reduction Optimization Algorithm for Ma...

引用

2024 IEEE Global Communications Conference, GLOBECOM 2024

作者： Li, Rengang Yan, Ruidong Guo, Zhenhua Qiu, Zhiyong Zhao, Yaqian Wang, Yanwei Tsinghua University Department of Computer Science and Technology Beijing China Inspur Electronic Information Industry Co. Ltd Beijing China

ISBN: (纸本)9798350351255

Nowadays, the variance reduction (VR) technique is used to improve the performance of gradient-type algorithms in machine learning and deep learning. However, some existing VR algorithms require unrealistic assumptions or conditions such as τ-gradient dominated and Polyak-Lojasiewicz (PL) conditions, which limit their applications. In this paper, we present a Double Mini-batch StochAstic Recursive grAdient algoritHm (DM-SARAH) without these assumptions or conditions to solve the convex and non-convex optimization problems respectively. The main contributions of this paper are twofold: (1) At the theoretical level, we optimize the convergence rate and provide a complexity analysis of DM-SARAH, and (2) At the experimental level, we evaluate the effectiveness and efficiency of the proposed algorithm on various datasets. The experimental results indicate that the proposed algorithm outperforms existing methods. © 2024 IEEE.

关键词： optimization algorithms

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：