检索结果-内蒙古大学图书馆

Enhancing multi-objective evolutionary neural architecture search with training-free Pareto local search

APPLIED INTELLIGENCE 2023年第8期53卷 8654-8672页

作者： Quan Minh Phan Ngoc Hoang Luong Univ Informat Technol Ho Chi Minh City Vietnam Vietnam Natl Univ Ho Chi Minh City Vietnam

Neural Architecture Search (NAS), that automates the design process of high-performing neural network architectures, is a multi-objective optimization problem. A single ideal architecture, that optimizes both predictive performance (e.g., the network accuracy) and computational costs (e.g., the model size, the number of parameters, the number of floating-point operations), does not exist. Instead, there is a Pareto front of multiple candidate architectures where each one represents an optimal trade-off between the competing objectives. Multi-Objective evolutionary Algorithms (MOEAs) are often employed to approximate such Pareto-optimal fronts for NAS problems. In this article, we introduce a local search method, namely Potential Solution Improving (PSI), that aims to improve certain potential solutions on approximation fronts to enhance the performance of MOEAs. The main bottleneck in NAS is the considerable computation cost that incurs from having to train a large number of candidate architectures to evaluate their accuracy. Recently, the Synaptic Flow has been proposed as a metric that relatively characterizes the performance of deep neural networks without running any training epoch. We thus propose that our PSI method can make use of this training-free metric as a proxy for network accuracy during local search steps. We conduct experiments with the well-known MOEA Non-dominated Sorting Genetic Algorithm II (NSGA-II) coupled with the training-free PSI local search in solving NAS problems created from the standard benchmarks NAS-Bench-101 and NAS-Bench-201. Experimental results confirm the efficiency enhancements brought about by our proposed method, which reduces the computational cost by four times compared to the baseline approach. The source code for the experiments in the article can be found at: https://***/ELO-Lab/MOENAS-TF-PSI.

关键词： Neural architecture search evolutionary computation Memetic algorithm Multi-objective optimization Synaptic flow

来源：评论

学校读者我要写书评

暂无评论

Design Exploration of Initial Bubbles for Higher-Speed Laser-Induced Microjets

引用

IEEE ACCESS 2023年 11卷 8268-8274页

作者： Ishikawa, Tatsumasa Nishida, Hiroyuki Tagawa, Yoshiyuki Tokyo Univ Agr & Technol Dept Mech Syst Engn Tokyo 1848588 Japan

Laser-induced microjets are advantageous for yielding high-speed and convergent microjets shape. However, owing to the need for high-power laser devices, reducing the required laser power is a prerequisite for convenience in practical use. This study aimed to optimize initial multiple bubbles design and find the dominant parameters for efficient microjet generation. For this purpose, the bubble design was optimized using an evolutionary algorithm-the covariance matrix adaptation evolutionary strategy. In addition, the dataset of solutions obtained through the optimization process was analyzed via principal component analysis (PCA). The optimization and analysis revealed that a higher microjet velocity could be obtained when a single bubble was generated near the tube wall, and an optimal total bubble volume was observed corresponding to this optimization. In addition, this approximate optimal solution we obtained was confirmed to reduce 88% of total initial energy compared to that of the typical solution. The result shows possibility to improve the efficiency of microjet generation and reducing the laser power.

关键词： Optimization computational modeling Mathematical models Principal component analysis Power lasers Liquids Viscosity Data visualization fluid dynamics evolutionary computation

来源：评论

学校读者我要写书评

暂无评论

An Improved Differential Evolution Framework Using Network Topology Information for Critical Nodes Detection

引用

IEEE TRANSACTIONS ON computationAL SOCIAL SYSTEMS 2023年第2期10卷 448-457页

作者： Yu, Shanqing Wang, Yongqi Li, Jiaxiang Fang, Xu Chen, Jinyin Zheng, Ziwan Fu, Chenbo Zhejiang Univ Technol Inst Cyberspace Secur Hangzhou 310023 Peoples R China Zhejiang Police Coll Hangzhou 310000 Peoples R China

Critical nodes detection (CND) focuses on identifying the nodes that significantly impact the network's robustness and is applied in various fields such as power grids, communication networks, and disease spreading. However, detecting the critical nodes is a challenging nondeterministic polynomial time complete (NP-complete) problem. One possible solution is using the evolutionary algorithm which has a high global search capability. However, the existing evolutionary algorithms for CND only focus on independent nodes, ignoring the underlying relationship among the nodes. Thus, in this work, we proposed a new topology-combined differential evolution framework called TDE to explore the possibility of improving the performance by fusing topology information, which designs individual genotypes through node degree, and new mutation and decoding-based selection operators are designed for these genotypes to use topology information effectively. The experiments on synthetic and real networks show that it is feasible to improve the search capability of the algorithm by fusing node degree information.

关键词： Network topology Topology evolutionary computation Approximation algorithms Decoding Measurement Encoding Complex network critical nodes detection (CND) differential evolution (DE) evolutionary computation

来源：评论

学校读者我要写书评

暂无评论

Intelligent Energy-Aware Routing Protocol in Mobile IoT Networks Based on SDN

IEEE TRANSACTIONS ON GREEN COMMUNICATIONS AND NETWORKING

引用

IEEE TRANSACTIONS ON GREEN COMMUNICATIONS AND NETWORKING 2023年第4期7卷 2093-2103页

作者： Samadi, Raheleh Nazari, Amin Seitz, Jochen Tech Univ Ilmenau Commun Networks Grp D-98693 Ilmenau Germany Bu Ali Sina Univ Artificial Intelligence Grp Hamadan *** Iran

Intelligent devices and equipment have affected almost all aspects of our life and behavior. The type of connection and the manner of communication between this large volume of devices has caused the emergence of a vast field in the Internet called the Internet of Things, which significantly highlights the issue of energy management and increases the lifetime of networks. Complex communications, especially in mobile networks, have generated many challenges for network designers. To solve these challenges, the Software Defined Networking (SDN) paradigm has reduced the overhead in the devices caused by processing and computing by adding new capabilities to mobile IoT networks. This technique transfers energy-consuming tasks to the central controller, which manages continuous topological changes of the network in dynamic environments. This paper presents a new routing approach called Intelligent Energy-aware Routing Protocol in Mobile IoT Networks based on SDN (IERMIoT), which tries to manage the dynamic changes of topology due to the movement of mobile nodes to increase the network's lifetime and prevent energy dissipation. For this purpose, it defines clusters of nodes and uses an intelligent evolutionary algorithm to determine the number of clusters required in the network and their balanced distribution in the dynamic environment. Also, this approach considers a mechanism to reduce the overhead of control packets and routing packets, which significantly affects the energy consumption of nodes. The simulation results indicate the proposed solution's effectiveness compared to other simulated approaches with respect to packet delivery rate, average energy consumption, network lifetime, number of alive nodes, coverage, and routing overhead.

关键词： Routing Internet of Things Clustering algorithms Wireless sensor networks Topology Sensors Network topology Software defined networking evolutionary computation evolutionary algorithm mobile Internet of Things routing

来源：评论

学校读者我要写书评

暂无评论

Dual-Fuzzy-Classifier-Based evolutionary Algorithm for Expensive Multiobjective Optimization

引用

IEEE TRANSACTIONS ON evolutionary computation 2023年第6期27卷 1575-1589页

作者： Zhang, Jinyuan He, Linjun Ishibuchi, Hisao Southern Univ Sci & Technol Dept Comp Sci & Engn Guangdong Prov Key Lab Brain Inspired Intelligent Shenzhen 518055 Peoples R China Natl Univ Singapore Dept Elect & Comp Engn Singapore 117575 Singapore

Multiobjective evolutionary algorithms (MOEAs) have been widely used to solve multiobjective optimization problems (MOPs). Conventional MOEAs usually require a large number of function evaluations (FEs) for evaluating the quality of solutions. However, only a limited number of FEs are affordable in many real-world optimization problems, where the FEs are computationally or economically expensive. The use of a large number of FEs decreases the effectiveness of MOEAs in solving these problems. This article proposes a dual-fuzzy-classifier-based surrogate model (DFC) and a DFC-based MOEA (DFC-MOEA) framework for expensive optimization problems. The DFC model is used for offspring preselection to choose high-quality offspring solutions and for categorization to divide unevaluated solutions into three categories. In the proposed framework, two fuzzy classifiers are built to predict the quality of each unevaluated solution. To strike a balance between convergence and diversity, one classifier is designed for predicting the dominance relation of unevaluated solutions (i.e., convergence), and another classifier is designed for predicting the crowdedness of unevaluated solutions (i.e., diversity). The integration of the proposed framework in three different types of MOEAs (i.e., a dominance-based MOEA, an indicator-based MOEA, and a decomposition-based MOEA) demonstrates its usefulness in handling expensive optimization problems. Comprehensive experiments on four well-known test suites and several real-world optimization problems demonstrate that the proposed framework is able to improve the performance of MOEAs under a limited number of FEs.

关键词： Optimization Iron Linear programming Convergence evolutionary computation computational modeling Training Expensive multiobjective optimization fuzzy classifiers multiobjective evolutionary algorithms (MOEAs) surrogate-based evolutionary algorithms

来源：评论

学校读者我要写书评

暂无评论

Optimization of dome shape for filament wound pressure vessels using data-driven evolutionary algorithms

引用

MATERIALS AND MANUFACTURING PROCESSES 2023年第15期38卷 1899-1910页

作者： Vondracek, Dominik Padovec, Zdenek Mares, Tomas Chakraborti, Nirupam Czech Tech Univ Fac Mech Engn Dept Mech Biomech & Mechatron Prague Czech Republic

Filament wound glass/epoxy and carbon/epoxy pressure vessels were designed using a data-driven evolutionary approach along with a classical mechanics-based analysis. Analytical solutions for three known end domes (a spherical shell, a geodesic-isotensoid shell and a shell based on minimizing of the Tsai-Hill's criterion) and end dome based on Hoffman's criterion were used to create surrogate models through two different evolutionary optimization algorithms: evolutionary Neural Net (EvoNN) and Bi-objective Genetic Programming (BioGP) and bi-objective evolutionary optimization studies were carried out using them. Composite pressure vessels are manufactured by means of filament (helical) winding. The stresses in the end domes were evaluated analytically and compared with each other. Hoffman strength criterion was chosen for the evaluation of the critical areas of the shells. Moreover, the best dome shape for given material and polar hole/equator ratio was selected for various types of failure.

关键词： Pressure vessels end dome helical winding composite material evolutionary computation genetic algorithms data-driven modeling multi-objective optimization EvoNN BioGP

来源：评论

学校读者我要写书评

暂无评论

Adaptive Caching Scheme for Jointly Optimizing Delay and Energy Consumption in Heterogeneous Digital Twin IoT

引用

IEEE TRANSACTIONS ON NETWORK SCIENCE AND ENGINEERING 2023年第6期10卷 4020-4032页

作者： Tan, Jingjing Tang, Fengxiao Zhao, Ming Zhu, Yusen Cent South Univ Sch Comp Sci & Engn Changsha 410075 Peoples R China Hunan Univ Sch Math Changsha 410012 Peoples R China

With the massive increase of mobile user and sensor data in the heterogeneous Internet of Things (IoT), the limited caching resources are challenging to meet people's increased data traffic demand. Due to ignoring the influence of uplink, the existing joint caching studies cannot reflect the actual network performance. Furthermore, some emerging delay-sensitive applications in the IoT, such as Virtual Reality, Augmented Reality, and autonomous driving, require better reliability and stability. Digital twins (DT) can provide real-time and stable wireless connectivity for heterogeneous IoT physical and virtual spaces. To address those issues, we propose an adaptive caching scheme based on the evolutionary game to jointly optimize delay and energy consumption in the uplink and downlink of heterogeneous digital twins IoT (HDT-IoT) networks. Besides, we prove the existence of evolutionary stability strategies (ESSs) in the proposed caching scheme. Based on the evolutionary stability analysis results, we derive the related expression between the content popularity and the ESS condition. The numerical results verify the existence of the content evolutionary stability caching strategy and the accuracy of the derived corresponding ESS conditional expressions and evaluate the caching performance by comparing our scheme with the other two caching schemes.

关键词： Internet of Things Delays Energy consumption Stability analysis Servers Games Digital twins Heterogeneous networks evolutionary computation heterogeneous digital twins IoT adaptive caching evolutionary game evolutionary stability strategies

来源：评论

学校读者我要写书评

暂无评论

Learning Symbolic Model-Agnostic Loss Functions via Meta-Learning

引用

IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE 2023年第11期45卷 13699-13714页

作者： Raymond, Christian Chen, Qi Xue, Bing Zhang, Mengjie Victoria Univ Wellington Sch Engn & Comp Sci Evolutionary Computat & Machine Learning Res Grp Wellington 6140 New Zealand

In this paper, we develop upon the emerging topic of loss function learning, which aims to learn loss functions that significantly improve the performance of the models trained under them. Specifically, we propose a new meta-learning framework for learning model-agnostic loss functions via a hybrid neuro-symbolic search approach. The framework first uses evolution-based methods to search the space of primitive mathematical operations to find a set of symbolic loss functions. Second, the set of learned loss functions are subsequently parameterized and optimized via an end-to-end gradient-based training procedure. The versatility of the proposed framework is empirically validated on a diverse set of supervised learning tasks. Results show that the meta-learned loss functions discovered by the newly proposed method outperform both the cross-entropy loss and state-of-the-art loss function learning methods on a diverse range of neural network architectures and datasets. We make our code available at *retracted*.

关键词： Task analysis Metalearning Optimization Training Market research Genetic programming Computer architecture Loss function learning meta-learning evolutionary computation neuro-symbolic

来源：评论

学校读者我要写书评

暂无评论

A Multi-Population Multi-Objective evolutionary Algorithm Based on the Contribution of Decision Variables to Objectives for Large-Scale Multi/Many-Objective Optimization

引用

IEEE TRANSACTIONS ON CYBERNETICS 2023年第11期53卷 6998-7007页

作者： Xu, Ying Xu, Chong Zhang, Huan Huang, Lei Liu, Yiping Nojima, Yusuke Zeng, Xiangxiang Hunan Univ Coll Comp Sci & Elect Engn Changsha 410082 Peoples R China Osaka Prefecture Univ Grad Sch Engn Sakai Osaka 5998531 Japan

Most existing multiobjective evolutionary algorithms treat all decision variables as a whole to perform genetic operations and optimize all objectives with one population at the same time. Considering different control attributes, different decision variables have different optimization effects on each objective, so decision variables can be divided into convergence-or diversity-related variables. In this article, we propose a new metric called the optimization degree of the convergence-related decision variable to each objective to calculate the contribution objective of each decision variable. All decision variables are grouped according to their contribution objectives. Then, a multiobjective evolutionary algorithm, namely, decision variable contributing to objectives evolutionary algorithm (DVCOEA), has been proposed. In order to balance the convergence and diversity of the population, the DVCOEA algorithm combines the multipopulation multiobjective framework, where two different optimization strategies are designed to optimize the subpopulation and individuals in the external archive, respectively. Finally, DVCOEA is compared with several state-of-the-art algorithms on a number of benchmark functions. Experimental results show that DVCOEA is a competitive approach for solving large-scale multi/many-objective problems.

关键词： Optimization evolutionary computation Convergence Statistics Sociology Diversity methods Clustering algorithms Decision variables analysis large-scale many-objective optimization large-scale multiobjective optimization multiobjective evolutionary algorithm multipopulation multiobjective evolutionary algorithm

来源：评论

学校读者我要写书评

暂无评论

Multiple objective optimisation for antenna diversity on airborne platforms

引用

IET SCIENCE MEASUREMENT & TECHNOLOGY 2023年第1期17卷 1-10页

作者： Emmett, Christian B. Flint, James A. Seager, Robert D. Loughborough Univ Wolfson Sch Mech Elect & Mfg Engn Loughborough Leics England

Vehicles such as automobiles, ships, satellites, and aircraft have a limited amount of physical space to install antennas for communications and navigation systems. This is exacerbated by the use of modern materials, like carbon fibre, and that large areas of the vehicles structure cannot be used to mount antenna, due to aerodynamic or other requirements. Therefore, it is necessary to be able to quickly and accurately find the optimum locations to mount a number of antenna systems, in a restricted space, whilst considering a number of different and sometimes contradictory antenna performance parameters. Thus, defining the optimum antenna locations is a multi-objective problem (MOP) and lends itself to the use of multi-objective evolutionary algorithms (MOEA). This paper presents a MOEA methodology that can be used to accurately, quickly, and robustly define the antenna locations. It will also define an appropriate MOEA and the fitness functions for predicting the radio frequency (RF) interoperability/mutual coupling between antenna systems and antenna RF radiation pattern installed performance.

关键词： genetic algorithms Optimisation techniques airborne platforms Pareto optimisation modern materials optimum antenna locations automobiles multiobjective problem evolutionary computation Antenna arrays aerodynamics optimum locations carbon fibre aerodynamic requirements restricted space Single antennas multiobjective evolutionary algorithms multiple objective optimisation antenna systems contradictory antenna performance parameters navigation systems physical space vehicles structure antenna RF radiation pattern installed performance antenna radiation patterns optimisation antenna diversity other requirements

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：