检索结果-内蒙古大学图书馆

An improved NSGA-III algorithm with adaptive mutation operator for big data optimization problems

FUTURE GENERATION COMPUTER SYSTEMS-THE INTERNATIONAL JOURNAL OF ESCIENCE 2018年 88卷 571-585页

作者： Yi, Jiao-Hong Deb, Suash Dong, Junyu Alavi, Amir H. Wang, Gai-Ge Qingdao Univ Technol Sch Informat & Control Engn Qingdao 266520 Peoples R China Victoria Univ Decis Sci & Modeling Program Melbourne Vic 8001 Australia Ocean Univ China Dept Comp Sci & Technol Qingdao 266100 Peoples R China Univ Missouri Dept Civil & Environm Engn Columbia MO 65211 USA Jiangsu Normal Univ Sch Comp Sci & Technol Xuzhou 221116 Jiangsu Peoples R China Northeast Normal Univ Inst Algorithm & Big Data Anal Changchun 130117 Jilin Peoples R China Northeast Normal Univ Sch Comp Sci & Informat Technol Changchun 130117 Jilin Peoples R China Jilin Univ Minist Educ Key Lab Symbol Computat & Knowledge Engn Changchun 130012 Jilin Peoples R China

One of the major challenges of solving big data optimization problems via traditional multi-objective evolutionary algorithms (MOEAs) is their high computational costs. This issue has been efficiently tackled by non -dominated sorting genetic algorithm, the third version, (NSGA-111). On the other hand, a concern about the NSGA-Ill algorithm is that it uses a fixed rate for mutation operator. To cope with this issue, this study introduces an adaptive mutation operator to enhance the performance of the standard NSGA-111 algorithm. The proposed adaptive mutation operator strategy is evaluated using three crossover operators of NSGA-111 including simulated binary crossover (SBX), uniform crossover (UC) and single point crossover (SI). Subsequently, three improved NSGA-111 algorithms (NSGA-111 SBXAM, NSGA-III SIAM, and NSGAIII UCAM) are developed. These enhanced algorithms are then implemented to solve a number of big data optimization problems. Experimental results indicate that NSGA-III with UC and adaptive mutation operator outperforms the other NSGA-111 algorithms. (C) 2018 Elsevier B.V. All rights reserved.

关键词： big data optimization Evolutionary multi-objective optimization NSGA-III Mutation operator Adaptive operators

来源：评论

学校读者我要写书评

暂无评论

jMetalSP: A framework for dynamic multi-objective big data optimization

引用

APPLIED SOFT COMPUTING 2018年 69卷 737-748页

作者： Barba-Gonzalez, Cristobal Garcia-Nieto, Jose Nebro, Antonio J. Cordero, Jose A. Durillo, Juan J. Navas-Delgado, Ismael Aldana-Montesa, Jose F. Univ Malaga Dept Lenguajes & Ciencias Comp Malaga Spain European Org Nucl Res CERN Geneva Switzerland Univ Innsbruck Distributed & Parallel Syst Grp Innsbruck Austria

Multi-objective metaheuristics have become popular techniques for dealing with complex optimization problems composed of a number of conflicting functions. Nowadays, we are in the big data era, so metaheuristics must be able to solve dynamic problems that may vary over time due to the processing and analysis of several streaming data sources. As this is a new field, there is a need for software platforms to solve dynamic multi-objective big data optimization problems. In this paper, we present jMetalSP, which combines the multi-objective optimization features of the jMetal framework with the streaming facilities of the Apache Spark cluster computing system. Thus, existing state-of-the-art multi-objective metaheuristics can be easily adapted to deal with dynamic optimization problems that are fed by multiple streaming data sources. Moreover, these algorithms can take advantage of the parallel computing features of Spark. We describe the architecture of jMetalSP and show how it can be used to solve a dynamic bi-objective instance of the Traveling Salesman Problem (TSP) based on New York City's real-time traffic data. We have also carried out an experimental study to assess the performance of the resultant jMetalSP application in a Hadoop cluster composed of 100 nodes. (C) 2017 Elsevier B.V. All rights reserved.

关键词： big data optimization Multi-objective optimization Metaheuristics Software framework jMetal Apache Spark

来源：评论

学校读者我要写书评

暂无评论

A comparative study between artificial bee colony (ABC) algorithm and its variants on big data optimization

引用

MEMETIC COMPUTING 2020年第2期12卷 129-150页

作者： Aslan, Selcuk Ondokuz Mayis Univ Samsun Turkey

The big data term and its formal definition have changed the properties of some of the computational problems. One of the problems for which the fundamental properties change with the existence of the big data is the optimization problems. Artificial bee colony (ABC) algorithm inspired by the intelligent source search, consumption and communication characteristics of the real honey bees has proven its efficiency on solving different numerical and combinatorial optimization problems. In this study, the standard ABC algorithm and its well-known variants including the gbest-guided ABC algorithm, the differential evolution based ABC/best/1 and ABC/best/2 algorithms, crossover ABC algorithm, converge-onlookers ABC algorithm and quick ABC algorithm were assessed using the electroencephalographic signal decomposition based optimization problems introduced at the 2015 Congress on Evolutionary Computing big data Competition. The experimental studies on solving big data optimization problems showed that the phase-divided structure of the standard ABC algorithm still protects its advantageous sides when the candidate food sources or solutions are generated by referencing the global best solution in the onlooker bee phase.

关键词： ABC algorithm big data optimization Signal decomposition

来源：评论

学校读者我要写书评

暂无评论

Differential evolution framework for big data optimization

引用

MEMETIC COMPUTING 2016年第1期8卷 17-33页

作者： Elsayed, Saber Sarker, Ruhul Univ New S Wales Sch Engn & Informat Technol Canberra ACT Australia Zagazig Univ Fac Comp & Informat Zagazig Egypt

During the last two decades, dealing with big data problems has become a major issue for many industries. Although, in recent years, differential evolution has been successful in solving many complex optimization problems, there has been research gaps on using it to solve big data problems. As a real-time big data problem may not be known in advance, determining the appropriate differential evolution operators and parameters to use is a combinatorial optimization problem. Therefore, in this paper, a general differential evolution framework is proposed, in which the most suitable differential evolution algorithm for a problem on hand is adaptively configured. A local search is also employed to increase the exploitation capability of the proposed algorithm. The algorithm is tested on the 2015 big data optimization competition problems (six single objective problems and six multi-objective problems). The results show the superiority of the proposed algorithm to several state-of-the-art algorithms.

关键词： Differential evolution big data optimization EEG signals

来源：评论

学校读者我要写书评

暂无评论

Fireworks algorithm framework for big data optimization

引用

MEMETIC COMPUTING 2016年第4期8卷 333-347页

作者： El Majdouli, Mohamed Amine Rbouh, Ismail Bougrine, Saad El Benani, Bouazza El Imrani, Abdelhakim Ameur Mohammed V Univ Fac Sci Concept & Syst Lab Rabat Morocco Mohammed V Univ Comp Sci Res Lab Fac Sci Rabat Morocco

This paper presents a novel optimization framework based on the Fireworks Algorithm for big data optimization problems. Indeed, the proposed framework is composed of two optimization algorithms. A single objective Fireworks Algorithm and a multi-objective Fireworks Algorithm are proposed for solving the big optimization of Signals problem "big-OPT" which belongs to the big data optimization problems class. The single objective Fireworks Algorithm adopts a modified search mechanism to ensure rapidity and preserve the explorative capacities of the basic Fireworks Algorithm. Afterward, the algorithm is extended to handle multi-objective optimization of big-OPT with a supplementary special sparks phase and a novel strategy for next generation selection. To validate the performance of the framework, extensive tests on six EEG datasets are performed. The framework is also compared with several approaches from recent state of the art. The study concludes the competitive performance of the proposed framework in comparison with the other techniques reported in this paper.

关键词： big data optimization Fireworks Algorithm big-OPT EEG Signals Brain Computer Interface

来源：评论

学校读者我要写书评

暂无评论

A genetic Artificial Bee Colony algorithm for signal reconstruction based big data optimization

引用

APPLIED SOFT COMPUTING 2020年第0期88卷 106053-000页

作者： Aslan, Selcuk Karaboga, Dervis Ondokuz Mayis Univ Dept Comp Engn Samsun Turkey Erciyes Univ Dept Comp Engn Kayseri Turkey King Abdulaziz Univ Fac Comp & Informat Technol Dept Informat Syst Jeddah Saudi Arabia

In recent years, the researchers have witnessed the changes or transformations driven by the existence of the big data on the definitions, complexities and future directions of the real world optimization problems. Analyzing the capabilities of the previously introduced techniques, determining possible drawbacks of them and developing new methods by taking into consideration of the unique properties related with the big data are nowadays in urgent demands. Artificial Bee Colony (ABC) algorithm inspired by the clever foraging behaviors of the real honey bees is one of the most successful swarm intelligence based optimization algorithms. In this study, a novel ABC algorithm based big data optimization technique was proposed. For exploring the solving abilities of the proposed technique, a set of experimental studies has been carried out by using different signal decomposition based big data optimization problems presented at the Congress on Evolutionary Computation (CEC) 2015 big data optimization Competition. The results obtained from the experimental studies first were compared with the well-known variants of the standard ABC algorithm named gbest-guided ABC (GABC), ABC/best/1, ABC/best/2, crossover ABC (CABC), converge-onlookers ABC (COABC) and quick ABC (qABC). The results of the proposed ABC algorithm were also compared with the Differential Evolution (DE) algorithm, Genetic algorithm (GA), Firefly algorithm (FA), Fireworks algorithm (FW), Phase Base optimization (PBO) algorithm, Particle Swarm optimization (PSO) algorithm and Dragonfly algorithm (DA) based big data optimization techniques. From the experimental studies, it was understood that the newly introduced ABC algorithm based technique is capable of producing better or at least promising results compared to the mentioned big data optimization techniques for all of the benchmark instances. (C) 2019 Elsevier B.V. All rights reserved.

关键词： big data optimization Signal decomposition Artificial Bee Colony

来源：评论

学校读者我要写书评

暂无评论

A hybrid multi-objective firefly algorithm for big data optimization

引用

APPLIED SOFT COMPUTING 2018年 69卷 806-815页

作者： Wang, Hui Wang, Wenjun Cui, Laizhong Sun, Hui Zhao, Jia Wang, Yun Xue, Yu Nanchang Inst Technol Jiangxi Prov Key Lab Water Informat Cooperat Sens Nanchang 330099 Jiangxi Peoples R China Nanchang Inst Technol Sch Informat Engn Nanchang 330099 Jiangxi Peoples R China Nanchang Inst Technol Sch Business Adm Nanchang 330099 Jiangxi Peoples R China Shenzhen Univ Coll Comp Sci & Software Engn Shenzhen 518060 Peoples R China Nanjing Univ Informat Sci & Technol Sch Comp & Software Nanjing 210044 Jiangsu Peoples R China

Multi-objective evolutionary algorithms (MOEAs) have shown good performance on many benchmark and real world multi-objective optimization problems. However, MOEAs may suffer from some difficulties when solving big data optimization problems with thousands of variables. Firefly algorithm (FA) is a new meta-heuristic, which has been proved to be a good optimization tool. In this paper, we present a hybrid multi-objective FA (HMOFA) for big data optimization. A set of big data optimization problems, including six single objective problems and six multi-objective problems, are tested in the experiments. Computational results show that HMOFA achieves promising performance on all test problems. (C) 2017 Elsevier B.V. All rights reserved.

关键词： Firefly algorithm (FA) Multi-objective firefly algorithm Multi-objective optimization big data optimization

来源：评论

学校读者我要写书评

暂无评论

Improved Particle Swarm optimization on Based Quantum Behaved Framework for big data optimization

引用

NEURAL PROCESSING LETTERS 2023年第3期55卷 2551-2586页

作者： Bas, Emine Selcuk Univ Kulu Vocat Sch TR-42075 Konya Turkey

In recent times, big data has become an essential concern with the rapid increase of digitalization. The problems that find solutions to the problems of finding and evaluating the features of big data are called optimization problems. In this paper, data sets containing EEG signals have been studied. The goal is to detect actual EEG signals while eliminating additional brain activity patterns in the collected data, resulting in more accurate interpretation. In the study, to handle big data optimization (big(Opt)) difficulties, a novel swarm intelligence-based technique is developed. A Developed PSO-Q was proposed by updating the random walking phase of the Particle Swarm optimization on the combined quantum behaved method (PSO-Q) for big(Opt) problems. PSO-Q's local search capability has been improved. The success of PSO-Q and IPSO-Q has been thoroughly tested in various cycles (maximum iterations) (300, 400, 500, and 1000) and population sizes (10, 25, and 50) on six data sets. The outcomes of the PSO-Q and IPSO-Q were statistically evaluated with the Wilcoxon Signed-Rank Test. PSO-Q and IPSO-Q have been compared with newly developed swarm-based algorithms (BA, Jaya, AOA, etc.) in the literature in recent years. The success of IPSO-Q has been shown by evaluating the results obtained. The results showed that IPSO-Q can be used as an alternative algorithm in big(Opt) problems.

关键词： big data optimization EEG signals Quantum-behaved particle swarm

来源：评论

学校读者我要写书评

暂无评论

Parallel coordinate descent methods for big data optimization

引用

MATHEMATICAL PROGRAMMING 2016年第1-2期156卷 433-484页

作者： Richtarik, Peter Takac, Martin Univ Edinburgh Sch Math Edinburgh Midlothian Scotland

In this work we show that randomized (block) coordinate descent methods can be accelerated by parallelization when applied to the problem of minimizing the sum of a partially separable smooth convex function and a simple separable convex function. The theoretical speedup, as compared to the serial method, and referring to the number of iterations needed to approximately solve the problem with high probability, is a simple expression depending on the number of parallel processors and a natural and easily computable measure of separability of the smooth component of the objective function. In the worst case, when no degree of separability is present, there may be no speedup;in the best case, when the problem is separable, the speedup is equal to the number of processors. Our analysis also works in the mode when the number of blocks being updated at each iteration is random, which allows for modeling situations with busy or unreliable processors. We show that our algorithm is able to solve a LASSO problem involving a matrix with 20 billion nonzeros in 2 h on a large memory node with 24 cores.

关键词： Parallel coordinate descent big data optimization Partial separability Huge-scale optimization Iteration complexity

来源：评论

学校读者我要写书评

暂无评论

Recent Advances in Randomized Methods for big data optimization

Recent Advances in Randomized Methods for Big Data Optimizat...

引用

作者： Jie Liu Lehigh University

学位级别：博士

In this thesis, we discuss and develop randomized algorithms for big data problems. In particular, we study the finite-sum optimization with newly emerged variance- reduction optimization methods (Chapter 2), explore the efficiency of second-order information applied to both convex and non-convex finite-sum objectives (Chapter 3) and employ the fast first-order method in power system problems (Chapter 4). In Chapter 2, we propose two variance-reduced gradient algorithms – mS2GD and SARAH. mS2GD incorporates a mini-batching scheme for improving the theoretical complexity and practical performance of SVRG/S2GD, aiming to minimize a strongly convex function represented as the sum of an average of a large number of smooth con- vex functions and a simple non-smooth convex regularizer. While SARAH, short for StochAstic Recursive grAdient algoritHm and using a stochastic recursive gradient, targets at minimizing the average of a large number of smooth functions for both con- vex and non-convex cases. Both methods fall into the category of variance-reduction optimization, and obtain a total complexity of O ((n + κ)log(1/ε)) to achieve an ε -accuracy solution for strongly convex objectives, while SARAH also maintains a sub-linear convergence for non-convex problems. Meanwhile, SARAH has a practical variant SARAH+ due to its linear convergence of the expected stochastic gradients in inner loops. In Chapter 3, we declare that randomized batches can be applied with second- order information, as to improve upon convergence in both theory and practice, with a framework of L-BFGS as a novel approach to finite-sum optimization problems. We provide theoretical analyses for both convex and non-convex objectives. Meanwhile, we propose LBFGS-F as a variant where Fisher information matrix is used instead of Hessian information, and prove it applicable to a distributed environment within the popular applications of least-square and cross-entropy losses. In Chapter 4, we develop fast rand

关键词： big data optimization Large-scale optimization Machine Learning Power Systems Randomized Methods

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：