检索结果-内蒙古大学图书馆

MOEA/D-ARA plus SBX: A new multi-objective evolutionary algorithm based on decomposition with artificial raindrop algorithm and simulated binary crossover

引用

KNOWLEDGE-BASED SYSTEMS 2016年 107卷 197-218页

作者： Jiang, Qiaoyong Wang, Lei Hei, Xinhong Yu, Guolin Lin, Yanyan Lu, Xiaofeng Xian Univ Technol Fac Comp Sci & Engn Xian 710048 Peoples R China Beifang Univ Nationalities Inst Appl Math Yinchuan 750021 Peoples R China

In the field of optimization computation, there has been a growing interest in applying intelligent algorithms to solve multi-objective optimization problems (MOPs). This paper focuses mainly on the multi objective evolutionary algorithm based on decomposition, MOEA/D for short, which offers a practical general algorithmic framework of evolutionary multi-objective optimization, and has been achieved great success for a wide range of MOPs. Like most other algorithms, however, MOEA/D has its limitations, which are reflected in three aspects: the problem of balancing diversity and convergence, non-uniform distribution of the Pareto front (PF), and weak convergence of the algorithm. To alleviate these limitations, a new combination of the artificial raindrop algorithm (ARA) and a simulated binary crossover (SBX) operator is first integrated into the framework of MOEA/D to balance the convergence and diversity. Thus, our proposed approach is called MOEA/D with ARA and SBX (MOEA/D-ARA+SBX). On the other hand, the raindrop pool in ARA is further extended to an external elitist archive, which retains only non-dominated solutions and discards all others. In addition, the k-nearest neighbors approach is introduced to prune away redundant non-dominated solutions. In such a way, a Pareto approximate subset with good distribution to the true PF may be achieved. Based on the relevant mathematical theory and some assumptions, it is proven that MOEA/D-ARA+SBX can converge to the true PF with probability one. For performance evaluation and comparison purposes, the proposed approach was applied to 44 multi-objective test problems with all types of Pareto set shape, and compared with 16 other versions of MOEA/D. The experimental results indicate its advantages over other approaches. (C) 2016 Elsevier B.V. All rights reserved.

关键词： multi-objective optimization Artificial raindrop algorithm Decomposition k-nearest neighbors External elitist archive Convergence analysis

来源：评论

学校读者我要写书评

暂无评论

Metaheuristics approaches to solve combinatorial optimization problems in distribution power systems. An application to Phase Balancing in low voltage three-phase networks

引用

INTERNATIONAL JOURNAL OF ELECTRICAL POWER & ENERGY SYSTEMS 2016年 76卷 1-10页

作者： Schweickardt, Gustavo Gimenez Alvarez, Juan Manuel Casanova, Carlos Univ Tecnol Nacl CONICET RA-3260 Concepcion Del Uruguay Argentina Univ Nacl San Juan CONICET Fac Ingn Desamparados San Juan Argentina

Metaheuristics algorithms are widely recognized as one of most practical approaches for combinatorial optimization problems. One the most interesting areas of application are the power systems. In particular, Distribution Systems planning and operation. This paper presents some metaheuristics approaches to solve a typical combinatorial optimization problem: the Phase Balancing in Low Voltage Electric Distribution Systems. A model supported in Linear Integer-Mixed Programming is presented, to observe and discussing its limitations. From this, is introduced a new metaheuristic, called Fuzzy Evolutionary Particle Swarm optimization, based in the Swarm Intelligence Principles and Evolution Strategies, which is extended to fuzzy domain to modeling a multi-objective optimization, by mean of a fuzzy fitness function. A simulation on a real system is presented, and advantages of this approach respect to the Classical Simulated Annealing and Particle Swarm metaheuristics, selected between the most representatives, are evidenced. (C) 2015 Elsevier Ltd. All rights reserved.

关键词： Metaheuristic algorithm Swarm Intelligence Fuzzy sets Electric distribution Phase Balancing

来源：评论

学校读者我要写书评

暂无评论

Neural network based multi objective optimization-a new algorithm 3rd

Neural network based multi objective optimization-a new algo...

引用

3rd International Conference on Foundations and Frontiers in Computer, Communication and Electrical Engineering (C2E2)

作者： Roy, Debasish Techno India Univ Kolkata India

ISBN: (纸本)9781315657912;9781138028777

The difference in Single objective and multi-objective optimizations is primarily in the cardinality in the optimal set;from a practical standpoint a user needs only one solution, no matter whether the associated optimization problem is single-objective or multi-objective. A multi-objective optimization problem can be converted into a single-objective optimization problem by the method of scalarizing an objective vector into a single composite objective function. This procedure of handling multi-objective optimization problems called Predilection Based MOO and is much simpler, yet still being more subjective than the any other procedure. In the evolutionary approaches the most popular methods are by evolutionary or genetic algorithms. This paper has proposed a new method for deriving Pareto Optimal Front by use of Neural Network. The algorithm has been presented here. A numerical experiment has also been performed to validate the algorithm.

关键词： multiobjective optimization

来源：评论

学校读者我要写书评

暂无评论

Expedited Simulation-Driven multi-objective Design optimization of Quasi-Isotropic Dielectric Resonator Antenna 1

引用

Conference on Simulation-Driven Modeling and optimization

作者： Bekasiewicz, Adrian Koziel, Slawomir Zieniutycz, Wlodzimierz Leifsson, Leifur Reykjavik Univ Sch Sci & Engn Engn Optimizat & Modeling Ctr Menntavegur 1 IS-101 Reykjavik Iceland Gdansk Univ Technol Fac Elect Telecommun & Informat PL-80233 Gdansk Poland Iowa State Univ Dept Aerosp Engn Ames IA 50011 USA

ISBN: (数字)9783319275178

ISBN: (纸本)9783319275178;9783319275154

Majority of practical engineering design problems require simultaneous handling of several criteria. Although many of design tasks can be turned into single-objective problems using sufficient formulations, in some situations, acquiring comprehensive knowledge about possible trade-offs between conflicting objectives may be necessary. This calls formulti-objective optimization that aims at identifying a set of alternative, Pareto-optimal designs. The most popular solution approaches to genuine multi-objective optimization include population-based metaheuristics. Unfortunately, such methods are not practical for problems involving expensive computational models, particularly for antenna engineering, where reliable design requires CPU-intensive electromagnetic (EM) analysis. In this work, we discuss two methodologies for expedited multi-objective design optimization of a six-parameter dielectric resonator antenna (DRA) with respect to three design criteria. The considered solution approaches rely on surrogate-based optimization (SBO) paradigm, where the design speedup is obtained by shifting the optimization burden into a cheap replacement model referred to as a surrogate. The latter is utilized for generating the initial approximation of the Pareto front representation as well as further refinement of the initially obtained Pareto-optimal solutions.

关键词： Computer-aided design (CAD) multi-objective optimization Antenna design Evolutionary algorithm Response surface approximation Kriging interpolation Design space reduction Dielectric resonator antenna

来源：评论

学校读者我要写书评

暂无评论

Pareto efficient multi-objective optimization for local tuning of analogy-based estimation

引用

NEURAL COMPUTING & APPLICATIONS 2016年第8期27卷 2241-2265页

作者： Azzeh, Mohammad Nassif, Ali Bou Banitaan, Shadi Almasalha, Fadi Appl Sci Univ Dept Software Engn POBOX 166 Amman Jordan Univ Sharjah Dept Elect & Comp Engn Sharjah U Arab Emirates Univ Detroit Mercy Dept Math Comp Sci & Software Engn Detroit MI 48221 USA Appl Sci Univ Dept Comp Sci POBOX 166 Amman Jordan

Analogy-based effort estimation (ABE) is one of the prominent methods for software effort estimation. The fundamental concept of ABE is closer to the mentality of expert estimation but with an automated procedure in which the final estimate is generated by reusing similar historical projects. The main key issue when using ABE is how to adapt the effort of the retrieved nearest neighbors. The adaptation process is an essential part of ABE to generate more successful accurate estimation based on tuning the selected raw solutions, using some adaptation strategy. In this study, we show that there are three interrelated decision variables that have great impact on the success of adaptation method: (1) number of nearest analogies (k), (2) optimum feature set needed for adaptation and (3) adaptation weights. To find the right decision regarding these variables, one need to study all possible combinations and evaluate them individually to select the one that can improve all prediction evaluation measures. The existing evaluation measures usually behave differently, presenting sometimes opposite trends in evaluating prediction methods. This means that changing one decision variable could improve one evaluation measure while it is decreasing the others. Therefore, the main theme of this research is how to come up with best decision variables that improve adaptation strategy and thus the overall evaluation measures without degrading the others. The impact of these decisions together has not been investigated before;therefore, we propose to view the building of adaptation procedure as a multi-objective optimization problem. The Particle swarm optimization algorithm (PSO) is utilized to find the optimum solutions for such decision variables based on optimizing multiple evaluation measures. We evaluated the proposed approaches over 15 datasets and using four evaluation measures. After extensive experimentation, we found that: (1) predictive performance of ABE has noticeably been

关键词： Analogy-based effort estimation Adaptation strategy Particle swarm optimization multi-objective optimization

来源：评论

学校读者我要写书评

暂无评论

Data allocation approaches for optimizing storage systems

Data allocation approaches for optimizing storage systems

引用

作者： Strong, Christina Rose University of California Santa Cruz

学位级别：Ph.D.

As storage systems grow in size and complexity, the necessity for automatically managing them increases. One area of concern is system optimization, often managed by the administrator manually tweaking parameters to make the system as efficient as possible. The problem arises when there are multiple objectives to optimize, some of which may interfere with each other. Even when optimization is automated, it often requires either a decision to be made as to which objectives are more important than others, or if the objectives are combined into a linear function, this function (and the potential related weights) must be determined by the administrator. To this end, most systems optimize for only one objective, or when there are two objectives it is two that do not conflict. However, when all objectives are equally important, or it is unknown which objective is most important, the system must learn to balance multiple objectives. One way to aid in automating system optimization is by looking at the file allocation problem. The file allocation problem considers a set of files (or tasks) that need to be allocated to some number of devices, so as to optimize an objective function. I propose extending the file allocation problem so that rather than optimizing a single objective function, the goal is to allocate data to a number of devices subject to a multi-objective optimization. This dissertation covers the three top-down theoretical approaches I developed as well as a bottom-up practical approach. In addition, I present a system design that incorporates the practical approach with two space saving modifications.

关键词：

来源：评论

学校读者我要写书评

暂无评论

multi-objective Parameter Configuration of Machine Learning Algorithms using Model-Based optimization

Multi-Objective Parameter Configuration of Machine Learning ...

引用

IEEE Symposium Series on Computational Intelligence (IEEE SSCI)

作者： Horn, Daniel Bischl, Bernd TU Dortmund Computat Stat D-44227 Dortmund Germany LMU Munchen Computat Stat D-80539 Munich Germany

ISBN: (纸本)9781509042401

The performance of many machine learning algorithms heavily depends on the setting of their respective hyperparameters. Many different tuning approaches exist, from simple grid or random search approaches to evolutionary algorithms and Bayesian optimization. Often, these algorithms are used to optimize a single performance criterion. But in practical applications, a single criterion may not be sufficient to adequately characterize the behavior of the machine learning method under consideration and the Pareto front of multiple criteria has to be considered. We propose to use model-based multi-objective optimization to efficiently approximate such Pareto fronts. Furthermore, the parameter space of many machine learning algorithms not only consists of numeric, but also categorical, integer or even hierarchical parameters, which is a general characteristic of many algorithm configuration tasks. Such mixed and hierarchical parameter space structures will be created naturally when one simultaneously performs model selection over a whole class of different possible prediction algorithms. Instead of tuning each algorithm individually, our approach can be used to configure machine learning pipelines with such a hierarchical structure, and efficiently operates on the joint space of all considered algorithms, in a multi-objective setting. Our optimization method is readily available as part of the mlrMBO R package on Github. We compare its performance against the TunePareto R package and the regular Latin Hypercube Sampling. We use a pure numerical setting of SVM parameter tuning and a mixed, hierarchical setting in which we optimize over multiple model spaces at once.

关键词： multiobjective optimization

来源：评论

学校读者我要写书评

暂无评论

multi-Antenna Transmission Using ESPAR with Peak Power Constraints

Multi-Antenna Transmission Using ESPAR with Peak Power Const...

引用

European Signal Processing Conference

作者： Lin Zhou Fahd Ahmed Khan Tharmalingam Ratnarajah Constantinos B. Papadias Institute for Digital Communications The University of Edinburgh Edinburgh UK School of Electrical Engineering and Computer Science (SEECS) National University of Science and Technology Pakistan Athens Information Technology (AIT) Athens Greece

Electronically steerable parasitic array radiator (ESPAR) technology provides multi-antenna transmission with a single radio frequency (RF) unit. In order to achieve stable transmission using an ESPAR antenna (EA), two approaches have been proposed in literature. One is to increase the self-resistance of an EA, the other is to transmit signals closely approximating the actual signals that keep the EA stable. In both approaches, no constraint on the transmission power of an EA was considered. This is not the case in actual systems, as the practical power amplifier normally has limited peak power. Taking into account the limited power availability, an optimization problem is formulated with the objective to minimize the MSE between the currents corresponding to the ideal and the approximate transmission signals. The non-convex problem is solved analytically by coordination transformation and a novel algorithm is proposed. It is shown that the system employing the proposed transmission scheme gives similar performance to that of a standard multiple antenna system, especially at low SNRs. In addition, it is shown that increasing the self-resistance of an EA to achieve stability is highly power inefficient.

关键词： optimization Transmitting antennas Radio frequency Impedance Resistance

来源：评论

学校读者我要写书评

暂无评论

Local-utopia Policy Selection for multi-objective Reinforcement Learning

Local-utopia Policy Selection for Multi-objective Reinforcem...

引用

IEEE Symposium Series on Computational Intelligence (IEEE SSCI)

作者： Parisi, Simone Blank, Alexander Viernicke, Tobias Peters, Jan Tech Univ Darmstadt Intelligent Autonomous Syst Grp D-64289 Darmstadt Germany Max Planck Inst Intelligent Syst D-72076 Tubingen Germany

ISBN: (纸本)9781509042401

Many real-world applications are characterized by multiple confticting objectives. In such problems, optimality is replaced by Pareto optimality and the goal is to find the Pareto frontier, a set 01' solutions representing different compromises among the objectives. Despite recent advances in multi-objective optimization, the selection, given the Pareto frontier, 01' a Pareto-optimal policy is still an important problem, prominent in practical applications such as economics and robotics. In this paper, we present a versatile approach for selecting a policy from the Pareto frontier according to user-defined preferences. Exploiting a novel scalarization function and heuristics, our approach provides an easy-to-use and effective method for Pareto-optimal policy selection. Furthermore, the scalarization is applicable in multiple-policy learning strategies tor approximating Pareto frontiers. To show the simplicity and effectiveness of our algorithm, we evaluate it on two problems and compare it to classical multi-objective reinforcement learning approaches.

关键词： multiobjective optimization

来源：评论

学校读者我要写书评

暂无评论

Recent Advances in Evolutionary multi-objective optimization

引用

2016年

作者： Slim Bechikh Rituparna Datta Abhishek Gupta

ISBN: (纸本)9783319429779

This book covers the most recent advances in the field of evolutionary multiobjective optimization. With the aim of drawing the attention of up-and coming scientists towards exciting prospects at the forefront of computational intelligence, the authors have made an effort to ensure that the ideas conveyed herein are accessible to the widest audience. The book begins with a summary of the basic concepts in multi-objective optimization. This is followed by brief discussions on various algorithms that have been proposed over the years for solving such problems, ranging from classical (mathematical) approaches to sophisticated evolutionary ones that are capable of seamlessly tackling practical challenges such as non-convexity, multi-modality, the presence of multiple constraints, etc. Thereafter, some of the key emerging aspects that are likely to shape future research directions in the field are presented. These include: optimization in dynamic environments, multi-objective bilevel programming, handling high dimensionality under many objectives, and evolutionary multitasking. In addition to theory and methodology, this book describes several real-world applications from various domains, which will expose the readers to the versatility of evolutionary multi-objective optimization.

关键词：

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：