检索结果-内蒙古大学图书馆

A multiobjective robust controller synthesis approach aided by multicriteria decision analysis

APPLIED SOFT COMPUTING 2017年 60卷 374-386页

作者： Gomes Bachur, Wagner Eustaquio Goncalves, Eduardo Nunes Ramirez, Jaime Arturo Batista, Lucas S. Univ Fed Minas Gerais Programa Posgrad Engn Eletr Av Antonio Carlos 6627 Belo Horizonte MG Brazil Ctr Fed Educ Tecnol Minas Gerais Dept Engn Eletr Av Amazonas 7675 Belo Horizonte MG Brazil Univ Fed Minas Gerais Operat Res & Complex Syst Lab ORCS Lab Dept Engn Eletr Av Antonio Carlos 6627 Belo Horizonte MG Brazil

This paper proposes a methodology for robust dynamic output-feedback control synthesis of uncertain linear systems represented by polytopic models. This control problem results in a semi-infinite optimization problem. The proposed synthesis procedure improves a previous two-step procedure, composed of synthesis and analysis, by employing both a multiobjective evolutionary algorithm (MOEA) and a multiple criteria decision making (MCDM) strategy in the synthesis step. The analysis stage is performed via a Branch and Bound (B&B) algorithm, enabling the validation of the former step. In the proposed multiobjective approach, the project aim is to meet the specifications of (i) reference signal tracking response, (ii) disturbance rejection and (iii) measurement noise attenuation. The project specifications are quantified in terms of H-infinity and H-2 norms of closed-loop transfer functions. Essentially, this technique combines the flexibility of a dedicated MOEA together with the minimax semi-infinite programming problem. Since a MOEA evolves a set of candidate solutions in parallel, the suggested strategy provides a diverse set of controller designs, which is very useful for an a posteriori decision-making process. A proposed multicriteria decision-aid strategy is employed, in addition, as a controller design tool, aiming to incorporate the decision-maker preferences throughout the process, which (i) enables a guided evolutionary search to a trade-off region (of solutions) of practical interest and (ii) assists the definition of an adequate final controller, characterized by a reasonable practical trade-off concerning the criteria. The application of the suggested framework is illustrated on three case studies in order to demonstrate the effectiveness of the method presented. (C) 2017 Elsevier B.V. All rights reserved.

关键词： Robust control Polytopic uncertainty multiobjective evolutionary optimization Semi-infinite programming Decision support

来源：评论

学校读者我要写书评

暂无评论

Enhanced Gradient for Differentiable Architecture Search

引用

IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS 2024年第7期35卷 9606-9620页

作者： Zhang, Haichao Hao, Kuangrong Gao, Lei Tang, Xuesong Wei, Bing Wei, Bing Donghua Univ Engn Res Ctr Digitized Text & Apparel Technol Minist Educ Shanghai 201620 Peoples R China Donghua Univ Coll Informat Sci & Technol Shanghai 201620 Peoples R China Commonwealth Sci & Ind Res Org CSIRO Waite Campus Urrbrae SA 5064 Australia

In recent years, neural architecture search (NAS) methods have been proposed for the automatic generation of task-oriented network architecture in image classification. However, the architectures obtained by existing NAS approaches are optimized only for classification performance and do not adapt to devices with limited computational resources. To address this challenge, we propose a neural network architecture search algorithm aiming to simultaneously improve network performance (e.g., classification accuracy) and reduce network complexity. The proposed framework automatically builds the network architecture at two stages: block-level search and network-level search. At the stage of block-level search, a relaxation method based on the gradient is proposed, using an enhanced gradient to design high-performance and low-complexity blocks. At the stage of network-level search, we apply an evolutionary multi-objective algorithm to complete the automatic design from blocks to the target network. The experiment results demonstrate that our method outperforms all evaluated hand-crafted networks in image classification, with an error rate of on CIFAR10 and an error rate of on CIFAR100, both at network parameter size less than one megabit. Moreover, compared with other neural architecture search methods, our method offers a tremendous reduction in designed network architecture parameters.

关键词： Convolutional neural networks (CNNs) evolutionary deep learning multiobjective evolutionary optimization neural architecture search (NAS)

来源：评论

学校读者我要写书评

暂无评论

Real-Time Federated evolutionary Neural Architecture Search

引用

IEEE TRANSACTIONS ON evolutionary COMPUTATION 2022年第2期26卷 364-378页

作者： Zhu, Hangyu Jin, Yaochu Univ Surrey Dept Comp Sci Guildford GU2 7XH Surrey England

Federated learning is a distributed machine learning approach to privacy preservation and two major technical challenges prevent a wider application of federated learning. One is that federated learning raises high demands on communication resources, since a large number of model parameters must be transmitted between the server and clients. The other challenge is that training large machine learning models such as deep neural networks in federated learning requires a large amount of computational resources, which may be unrealistic for edge devices such as mobile phones. The problem becomes worse when deep neural architecture search (NAS) is to be carried out in federated learning. To address the above challenges, we propose an evolutionary approach to real-time federated NAS that not only optimizes the model performance but also reduces the local payload. During the search, a double-sampling technique is introduced, in which for each individual, only a randomly sampled submodel is transmitted to a number of randomly sampled clients for training. This way, we effectively reduce computational and communication costs required for evolutionary optimization, making the proposed framework well suitable for real-time federated NAS.

关键词： Computational modeling Computer architecture Servers Real-time systems Collaborative work Training optimization AutoML communication cost deep neural networks (DNNs) federated learning multiobjective evolutionary optimization neural architecture search real-time optimization

来源：评论

学校读者我要写书评

暂无评论

Multiclass Gene Selection Using Pareto-Fronts

引用

IEEE-ACM TRANSACTIONS ON COMPUTATIONAL BIOLOGY AND BIOINFORMATICS 2013年第1期10卷 87-97页

作者： Rajapakse, Jagath C. Mundra, Piyushkumar A. Nanyang Technol Univ Sch Comp Engn BioInformat Res Ctr Singapore 639798 Singapore MIT Dept Biol Engn Cambridge MA 02139 USA Singapore Alliance Singapore Singapore

Filter methods are often used for selection of genes in multiclass sample classification by using microarray data. Such techniques usually tend to bias toward a few classes that are easily distinguishable from other classes due to imbalances of strong features and sample sizes of different classes. It could therefore lead to selection of redundant genes while missing the relevant genes, leading to poor classification of tissue samples. In this manuscript, we propose to decompose multiclass ranking statistics into class-specific statistics and then use Pareto-front analysis for selection of genes. This alleviates the bias induced by class intrinsic characteristics of dominating classes. The use of Pareto-front analysis is demonstrated on two filter criteria commonly used for gene selection: F-score and KW-score. A significant improvement in classification performance and reduction in redundancy among top-ranked genes were achieved in experiments with both synthetic and real-benchmark data sets.

关键词： Aggregation statistics filter methods gene selection multiobjective evolutionary optimization Pareto-front analysis

来源：评论

学校读者我要写书评

暂无评论

Design of Hybrid Distillation/Melt Crystallization Processes for Separation of Close-Boiling Mixtures

引用

CHEMIE INGENIEUR TECHNIK 2012年第11期84卷 2035-2047页

作者： Micovic, Jovana Beierling, Thorsten Lutze, Philip Sadowski, Gabriele Gorak, Andrzej Tech Univ Dortmund Lehrstuhl Fluidverfahrenstech Emil Figge Str 70 D-44227 Dortmund Germany Tech Univ Lodz PL-93005 Lodz Poland TU Dortmund Lehrstuhl Thermodynam D-44227 Dortmund Germany

Design of hybrid separations is still challenging due to large number of structural and operation degrees of freedom. To design a hybrid separation combining distillation and crystallization, a three-step approach is utilized, which can only be applied if all parameters for the rigorous modeling of crystallization and all cost functions are known a priori. Since this is often not the case in early process development stages, a four-step design method is proposed for that purpose. The developed tools and the feasibility of the approach are shown in a case study of separation of a binary mixture of long-chain isomeric aldehydes.

关键词： multiobjective evolutionary optimization Hybrid separation processes Long-chain isomeric aldehydes Melt crystallization Modeling

来源：评论

学校读者我要写书评

暂无评论

A multiobjective evolutionary Algorithm for Finding Knee Regions Using Two Localized Dominance Relationships

引用

IEEE TRANSACTIONS ON evolutionary COMPUTATION 2021年第1期25卷 145-158页

作者： Yu, Guo Jin, Yaochu Olhofer, Markus Univ Surrey Dept Comp Sci Guildford GU2 7XH Surrey England Honda Res Inst Europe GmbH D-63073 Offenbach Germany

In preference-based optimization, knee points are considered the naturally preferred tradeoff solutions, especially when the decision maker has little a priori knowledge about the problem to be solved. However, identifying all convex knee regions of a Pareto front remains extremely challenging, in particular in a high-dimensional objective space. This article presents a new evolutionary multiobjective algorithm for locating knee regions using two localized dominance relationships. In the environmental selection, the alpha-dominance is applied to each subpopulation partitioned by a set of predefined reference vectors, thereby guiding the search toward different potential knee regions while removing possible dominance resistant solutions. A knee-oriented-dominance measure making use of the extreme points is then proposed to detect knee solutions in convex knee regions and discard solutions in concave knee regions. Our experimental results demonstrate that the proposed algorithm outperforms the state-of-the-art knee identification algorithms on a majority of multiobjective optimization test problems having up to eight objectives and a hybrid electric vehicle controller design problem with seven objectives.

关键词： alpha-dominance knee-oriented dominance knees multiobjective evolutionary optimization preference

来源：评论

学校读者我要写书评

暂无评论

ASMiGA: An Archive-Based Steady-State Micro Genetic Algorithm

引用

IEEE TRANSACTIONS ON CYBERNETICS 2015年第1期45卷 40-52页

作者： Nag, Kaustuv Pal, Tandra Pal, Nikhil R. Jadavpur Univ Dept Instrumentat & Elect Engn Kolkata 700108 India Natl Inst Technol Dept CSE Durgapur 713209 India Indian Stat Inst Elect & Commun Sci Unit Kolkata 700108 India

We propose a new archive-based steady-state micro genetic algorithm (ASMiGA). In this context, a new archive maintenance strategy is proposed, which maintains a set of nondominated solutions in the archive unless the archive size falls below a minimum allowable size. It makes the archive size adaptive and dynamic. We have proposed a new environmental selection strategy and a new mating selection strategy. The environmental selection strategy reduces the exploration in less probable objective spaces. The mating selection increases searching in more probable search regions by enhancing the exploitation of existing solutions. A new crossover strategy DE-3 is proposed here. ASMiGA is compared with five well-known multiobjective optimization algorithms of different types-generational evolutionary algorithms (SPEA2 and NSGA-II), archive-based hybrid scatter search, decomposition-based evolutionary approach, and archive-based micro genetic algorithm. For comparison purposes, four performance measures (HV, GD, IGD, and GS) are used on 33 test problems, of which seven problems are constrained. The proposed algorithm outperforms the other five algorithms.

关键词： Archive-based algorithm genetic algorithms multiobjective evolutionary optimization Pareto front

来源：评论

学校读者我要写书评

暂无评论

A multiobjective study of the Gaussian cluster paging in the Reporting Cells strategy

引用

APPLIED SOFT COMPUTING 2015年 28卷 332-344页

作者： Berrocal-Plaza, Victor Vega-Rodriguez, Miguel A. Sanchez-Perez, Juan M. Univ Extremadura Dept Comp & Commun Technol Escuela Politecn Caceres 10003 Spain

Mobile location management is an important task in mobile networks. This paper presents an analysis of the Gaussian cluster paging in the Reporting Cells strategy. Reporting Cells define a location update strategy in which some network cells are configured as Reporting Cells with the aim of tracking the subscribers' movement. On the other hand, the Gaussian cluster paging is a paging scheme where we assume that the probability of finding a callee's terminal decreases as we move away from the last updated location. This paging scheme is studied in a multiobjective way for different probability thresholds, and considering time-delay constraints. With the goal of finding the best possible sets of non-dominated solutions, we propose our implementation of a well-known multiobjective evolutionary algorithm: the Non-dominated Sorting Genetic Algorithm II. By means of an experimental study, we show that each probability threshold has its own non-dominated region in the objective space. (C) 2014 Elsevier B.V. All rights reserved.

关键词： Mobile location management Reporting Cells scheme Gaussian paging multiobjective evolutionary optimization Non-dominated Sorting Genetic Algorithm II

来源：评论

学校读者我要写书评

暂无评论

Optimizing the mobility management task in networks of four world capital cities

引用

JOURNAL OF NETWORK AND COMPUTER APPLICATIONS 2015年 51卷 18-28页

作者： Berrocal-Plaza, Victor Vega-Rodriguez, Miguel A. Sanchez-Perez, Juan M. Univ Extremadura Escuela Politecn Dept Technol Comp & Commun Caceres 10003 Spain

In this paper, we study one of the most important management tasks in any Public Land Mobile Network: the mobile location management. In this management task, the network cells are grouped into logical areas (called the Registration Areas) in order to avoid the overload of the signaling network. The proper dimensioning of these Registration Areas is an important engineering issue that is modeled as a bi-objective optimization problem. In previous works by other authors, the Registration Areas planning problem was optimized by using the linear aggregation of the objective functions. This technique allows simplifying the optimization problem but has several drawbacks. That is why, we use our versions of two multiobjective optimization algorithms. Furthermore, a multiobjective approach provides a wide range of solutions among which the network operator could select the one that best meets its real requirements. With the aim of studying the mobile location management in realistic mobile environments, we have generated four novel mobile activity traces by importing four real networks topologies into a mobile networks simulation tool. Experimental results show that our proposals are very interesting because they achieve better solutions than the single-objective optimization algorithms proposed by other authors in a very less execution time. (C) 2015 Elsevier Ltd. All rights reserved.

关键词： Mobile location management Registration areas planning multiobjective evolutionary optimization

来源：评论

学校读者我要写书评

暂无评论

Preference-based Pareto optimization in certain and noisy environments

引用

ENGINEERING optimization 2009年第1期41卷 23-38页

作者： Trautmann, Heike Mehnen, Joern TU Dortmund Univ Stat Fac D-44221 Dortmund Germany Cranfield Univ Decis Engn Ctr Cranfield MK43 0AL Beds England

In this article a method for including a priori preferences of decision makers into multicriteria optimization problems is presented. A set of Pareto-optimal solutions is determined via desirability functions of the objectives which reveal experts' preferences regarding different objective regions. An application to noisy objective functions is not straightforward but very relevant for practical applications. Two approaches are introduced in order to handle the respective uncertainties by means of the proposed preference-based Pareto optimization. By applying the methods to the original and uncertain Binh problem and a noisy single cut turning cost optimization problem, these approaches prove to be very effective in focusing on different parts of the Pareto front of the ori-ginal problem in both certain and noisy environments.

关键词： multiobjective evolutionary optimization multiobjective evolutionary algorithm noisy environment desirability function single cut turning problem

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：