检索结果-内蒙古大学图书馆

A new real-coded stochastic Bayesian optimization algorithm for continuous global optimization

GENETIC PROGRAMMING AND EVOLVABLE MACHINES 2016年第2期17卷 145-167页

作者： Moradabadi, Behnaz Ebadzadeh, Mohammad Mahdi Meybodi, Mohammad Reza Amirkabir Univ Technol Dept Comp Engn Tehran Iran

estimation of distribution algorithms are considered to be a new class of evolutionary algorithms which are applied as an alternative to genetic algorithms. Such algorithms sample the new generation from a probabilistic model of promising solutions. The search space of the optimization problem is improved by such probabilistic models. In the Bayesian optimization algorithm (BOA), the set of promising solutions forms a Bayesian network and the new solutions are sampled from the built Bayesian network. This paper proposes a novel real-coded stochastic BOA for continuous global optimization by utilizing a stochastic Bayesian network. In the proposed algorithm, the new Bayesian network takes advantage of using a stochastic structure (that there is a probability distribution function for each edge in the network) and the new generation is sampled from the stochastic structure. In order to generate a new solution, some new structure, and therefore a new Bayesian network is sampled from the current stochastic structure and the new solution will be produced from the sampled Bayesian network. Due to the stochastic structure used in the sampling phase, each sample can be generated based on a different structure. Therefore the different dependency structures can be preserved. Before the new generation is generated, the stochastic network's probability distributions are updated according to the fitness evaluation of the current generation. The proposed method is able to take advantage of using different dependency structures through the sampling phase just by using one stochastic structure. The experimental results reported in this paper show that the proposed algorithm increases the quality of the solutions on the general optimization benchmark problems.

关键词： Evolutionary algorithms estimation of distribution algorithms Bayesian optimization algorithms Bayesian networks

来源：评论

学校读者我要写书评

暂无评论

PASCAL: An EDA for Parameterless Shape-Independent Clustering

PASCAL: An EDA for Parameterless Shape-Independent Clusterin...

引用

IEEE Congress on Evolutionary Computation (CEC) held as part of IEEE World Congress on Computational Intelligence (IEEE WCCI)

作者： Cagnini, Henry E. L. Barros, Rodrigo C. Pontificia Univ Catolica Rio Grande do Sul Porto Alegre RS Brazil

ISBN: (纸本)9781509006229

Data clustering is an unsupervised learning task that can be regarded as the combinatorial optimisation NP-hard problem of assigning N objects to one (or more) among k clusters. Most data clustering algorithms require the user to set a number of pre-defined parameters that have decisive impact in the formation of clusters, such as the number of clusters (initial or final), cluster radius, minimum number of objects, and similar parameters. In addition, several clustering algorithms are limited with regard to the shape of clusters that can be found, a limitation usually resulting from the optimisation process performed over a given distance metric. In this work, we propose a novel clustering algorithm that addresses the two aforementioned problems regarding the amount of parameters and cluster shape. Our approach makes use of the theory of estimation of distribution algorithms in order to probabilistic sample a set of must-link/cannot link constraints in order to generate a data partition with the proper number of clusters. We name our method PASCAL, and we empirically show that it is capable of not only detecting the right number of clusters but also of properly assigning objects to the correct cluster in a variety of artificial and real problems whose solutions are known in advance.

关键词： estimation of distribution algorithms clustering machine learning

来源：评论

学校读者我要写书评

暂无评论

On the Design of Hard mUBQP Instances 16

On the Design of Hard mUBQP Instances

引用

Genetic and Evolutionary Computation Conference (GECCO)

作者： Zangari, Murilo Santana, Roberto Mendiburu, Alexander Pozo, Aurora Univ Fed Parana Curitiba Parana Brazil Univ Basque Country UPV EHU San Sebastian Spain

ISBN: (纸本)9781450342063

This paper proposes a new method for the design and analysis of multi-objective unconstrained binary quadratic programming (mUBQP) instances, commonly used for testing discrete multi-objective evolutionary algorithms (MOEAs). These instances are usually generated considering the sparsity of the matrices and the correlation between objectives but randomly selecting the values for the matrix cells. Our hypothesis is that going beyond the use of objective correlations by considering different types of variables interactions in the generation of the instances can help to obtain more diverse problem benchmarks, comprising harder instances. We propose a parametric approach in which small building blocks of deceptive functions are planted into the matrices that define the mUBQP. The algorithm for creating the new instances is presented, and the difficulty of the functions is tested using variants of a decomposition-based MOEA. Our experimental results confirm that the instances generated by planting deceptive blocks require more function evaluations to be solved than instances generated using other methods.

关键词： deceptive functions MOEA/D multi-objective optimization multi-objective unconstrained binary quadratic programming PBIL estimation of distribution algorithms

来源：评论

学校读者我要写书评

暂无评论

An Improved DBOA Based on estimation of Model Similarity 7th

An Improved DBOA Based on Estimation of Model Similarity

引用

7th International Symposium on Intelligence Computation and Applications (ISICA)

作者： Kong, Yuyan Yao, Jintao Yang, Lei Guangzhou Coll Commerce Guangzhou 511363 Guangdong Peoples R China South China Agr Univ Coll Math & Informat Guangzhou 510642 Guangdong Peoples R China

ISBN: (纸本)9789811003561;9789811003554

In DBOA, to build accurately the best Bayesian network with respect to most metrics is NP-complete and the high time complexity of learning the model structure becomes a bottleneck of DBOA for real application. Consequently, in order to decrease the asymptotic time complexity of model building and make the algorithm more practical even for extremely large and complex problem, this paper presents adaptive sporadic model building based on estimation of model similarity as an efficiency enhancement technique of DBOA. The results show that performing the adaptive model building in DBOA can reduce the number of building model under no increasing on the number of generation and population size necessary to converge to optimal solutions, and achieve a better trade-off between the convergence speed and convergence results.

关键词： estimation of distribution algorithms Bayesian optimization algorithm Bayesian networks Model similarity Adaptive model building

来源：评论

学校读者我要写书评

暂无评论

HYBRID MUSIC RECOMMENDER USING CONTENT-BASED AND SOCIAL INFORMATION 41

HYBRID MUSIC RECOMMENDER USING CONTENT-BASED AND SOCIAL INFO...

引用

41st IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)

作者： Chiliguano, Paulo Fazekas, Gyorgy Queen Mary Univ London Sch Elect Engn & Comp Sci Mile End Rd London E1 4NS England

ISBN: (纸本)9781479999880

Internet resources available today, including songs, albums, playlists or podcasts, that a user cannot discover if there is not a tool to filter the items that the user might consider relevant. Several recommendation techniques have been developed since the Internet explosion to achieve this filtering task. In an attempt to recommend relevant songs to users, we propose an hybrid recommender that considers real-world users information and high-level representation for audio data. We use a deep learning technique, convolutional deep neural networks, to represent an audio segment in a n-dimensional vector, whose dimensions define the probability of the segment to belong to a specific music genre. To capture the listening behavior of a user, we investigate a state-of-the-art technique, estimation of distribution algorithms. The designed hybrid music recommender outperforms the predictions compared with a traditional content-based recommender.

关键词： Deep Learning Convolutional Neural Networks estimation of distribution algorithms Recommender Systems MIR

来源：评论

学校读者我要写书评

暂无评论

Investigating selection strategies in multi-objective probabilistic model based algorithms 5

Investigating selection strategies in multi-objective probab...

引用

5th Brazilian Conference on Intelligent Systems (BRACIS)

作者： Strickler, Andrei Castro, Olacir, Jr. Pozo, Aurora Santana, Roberto Univ Fed Parana Comp Sci Dept Curitiba Parana Brazil Univ Basque Country Dept Comp Sci & Artificial Intelligence San Sebastian 20080 Spain

ISBN: (纸本)9781509035663

Recent advances on multi-objective evolutionary algorithm (MOEAs) have acknowledged the important role played by selection, replacement, and archiving strategies in the behavior of these algorithms. However, the influence of these methods has been scarcely investigated for the particular class of MOEAs that use probabilistic modeling of the solutions. In this paper we fill this void by proposing an analysis of the role of the aforementioned strategies on an extensive set of bi-objective functions. We focus on the class of algorithms that use Gaussian univariate marginal models, and study how typical selection and replacement strategies used together with this probabilistic model impact the behavior of the search. Our analysis is particularized for a set of bi-objective functions that exhibit a representative set of characteristics (e.g. decomposable, ill-conditioned, non-linear, etc.). The experimental results shows that MOEAs that use simple probabilistic modeling outperform traditional MOEAs based on crossover operators.

关键词： multi-objective optimization estimation of distribution algorithms Gaussian UMDA selection

来源：评论

学校读者我要写书评

暂无评论

estimation of distribution algorithm with scatter search for dynamic optimisation problems

引用

INTERNATIONAL JOURNAL OF COMPUTING SCIENCE AND MATHEMATICS 2015年第3期6卷 221-231页

作者： Yu, Fahong He, Feng Chen, Meijia Ma, Longhua Lu, Zheming Jiaxing Univ Coll Math & Informat Engn Jiaxing Zhejiang Peoples R China Zhejiang Univ Ningbo Inst Technol Jiaxing Zhejiang Peoples R China Jiaxing Univ Ctr Econ Managing Expt Jiaxing Zhejiang Peoples R China Zhejiang Univ Sch Aeronaut & Astronaut Hangzhou Zhejiang Peoples R China

Aiming at the trouble to track the optima in dynamic environments with estimation of distribution algorithms (EDAs). An estimation of distribution algorithm with scatter search (EDASS) is proposed in this paper. Its basic idea is to employ a scatter search to increase the diversity in a guided fashion and an adaptive leader clustering method to locate multiple local optima. Both the information of current population and the part history information were referred for building probability model. The experimental results show that the EDASS is effective for dynamic optimisation problems.

关键词： estimation of distribution algorithms EDAs dynamic optimisation problems DOPs clustering scatter search

来源：评论

学校读者我要写书评

暂无评论

estimation of distribution algorithms For Job Schedule Problem

Estimation of Distribution Algorithms For Job Schedule Probl...

引用

2nd International Conference on Information and Computing Science

作者： Li, Jun Jiang, Yong Nanjing Univ Sci & Technol Dept Informat & Comp Sci Nanjing 210094 Peoples R China

ISBN: (纸本)9780769536347

Well known job schedule problem is NP-hard in management and operating system. A few of evolution algorithms such as GA have been proposed to approach this kind of problems. At present estimation of distribution algorithms are the leading heuristic algorithms which overcome the defect of tradition GA and reflex more excellent property than others. In this paper effective of EDA are tested by the job scheduling problem and the simulation results are analyzed in details.

关键词： estimation of distribution algorithms job schedule problem

来源：评论

学校读者我要写书评

暂无评论

An overview of population-based algorithms for multi-objective optimisation

引用

INTERNATIONAL JOURNAL OF SYSTEMS SCIENCE 2015年第9期46卷 1572-1599页

作者： Giagkiozis, Ioannis Purshouse, Robin C. Fleming, Peter J. Univ Sheffield Dept Automat Control & Syst Engn Sheffield S1 3JD S Yorkshire England

In this work we present an overview of the most prominent population-based algorithms and the methodologies used to extend them to multiple objective problems. Although not exact in the mathematical sense, it has long been recognised that population-based multi-objective optimisation techniques for real-world applications are immensely valuable and versatile. These techniques are usually employed when exact optimisation methods are not easily applicable or simply when, due to sheer complexity, such techniques could potentially be very costly. Another advantage is that since a population of decision vectors is considered in each generation these algorithms are implicitly parallelisable and can generate an approximation of the entire Pareto front at each iteration. A critique of their capabilities is also provided.

关键词： ant colony optimisation artificial immune systems differential evolution particle swarm optimisation genetic algorithms estimation of distribution algorithms

来源：评论

学校读者我要写书评

暂无评论

Improved sampling using loopy belief propagation for probabilistic model building genetic programming

引用

SWARM AND EVOLUTIONARY COMPUTATION 2015年 23卷 1-10页

作者： Sato, Hiroyuki Hasegawa, Yoshihiko Bollegala, Danushka Iba, Hitoshi Univ Tokyo Grad Sch Engn Tokyo 1138654 Japan Univ Tokyo Grad Sch Informat Sci & Technol Tokyo 1138654 Japan Univ Liverpool Dept Comp Sci Liverpool L69 3BX Merseyside England

In recent years, probabilistic model building genetic programming (PMBGP) for program optimization has attracted considerable interest. PMBGPs generally use probabilistic logic sampling (PLS) to generate new individuals. However, the generation of the most probable solutions (MPSs), i.e., solutions with the highest probability, is not guaranteed. In the present paper, we introduce loopy belief propagation (LBP) for PMBGPs to generate MPSs during the sampling process. We selected program optimization with linkage estimation (POLE) as the foundation of our approach and we refer to our proposed method as POLE-BP. We apply POLE-BP and existing methods to three benchmark problems to investigate the effectiveness of LBP in the context of PMBGPs, and we describe detailed examinations of the behaviors of LBP. We find that POLE-BP shows better search performance with some problems because LBP boosts the generation of building blocks. (C) 2015 Elsevier B.V. All rights reserved.

关键词： Genetic programming estimation of distribution algorithms Loopy belief propagation Probabilistic model building GP

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：