Cooperative Metaheuristics for Exploring Proteomic Data

Most combinatorial optimization problems cannotbe solved exactly. A class of methods, calledmetaheuristics, has proved its efficiency togive good approximated solutions in areasonable time. Cooperative metaheuristics area sub-set of metaheuristics, which implies aparallel exploration of the search space byseveral entities with information exchangebetween them. The importance of informationexchange in the optimization process is relatedto the building block hypothesis ofevolutionary algorithms, which is based onthese two questions: what is the pertinentinformation of a given potential solution andhow this information can be shared? Aclassification of cooperative metaheuristicsmethods depending on the nature of cooperationinvolved is presented and the specificproperties of each class, as well as a way tocombine them, is discussed. Severalimprovements in the field of metaheuristics arealso given. In particular, a method to regulatethe use of classical genetic operators and todefine new more pertinent ones is proposed,taking advantage of a building block structuredrepresentation of the explored space. Ahierarchical approach resting on multiplelevels of cooperative metaheuristics is finallypresented, leading to the definition of acomplete concerted cooperation strategy. Someapplications of these concepts to difficultproteomics problems, including automaticprotein identification, biological motifinference and multiple sequence alignment arepresented. For each application, an innovativemethod based on the cooperation concept isgiven and compared with classical approaches.In the protein identification problem, a firstlevel of cooperation using swarm intelligenceis applied to the comparison of massspectrometric data with biological sequencedatabase, followed by a genetic programmingmethod to discover an optimal scoring function.The multiple sequence alignment problem isdecomposed in three steps involving severalevolutionary processes to infer different kindof biological motifs and a concertedcooperation strategy to build the sequencealignment according to their motif content.     

Using machine learning in a cooperative hybrid parallel strategy of metaheuristics

This paper proposes the construction of a centralized hybrid metaheuristic cooperative strategy to solve optimization problems. Knowledge (intelligence) is incorporated into the coordinator to improve performance. This knowledge is incorporated through a set of rules and models obtained from a knowledge extraction process applied to the records of the results returned by individual metaheuristics. The effectiveness of the approach is tested in several computational experiments in which we compare the results obtained by the individual metaheuristics, by several non-cooperative and cooperative strategies and by the strategy proposed in this paper.     

A cooperation strategy based on bargaining solution in wireless sensor networks

This paper presents a symmetric cooperation strategy for wireless sensor networks, aiming to improve the transmission efficiency of the network. The cooperation strategy is implemented by partitioning the nodes into several cooperative groups. Then, in each group, the optimal cooperative bandwidth allocation is obtained based on Raiffa-KalaiSmorodinsky bargaining solution (RBS). Numerical results show that the symmetric cooperation strategy can improve the sensor node’s transmission efficiency dramatically.     

A hybrid search combining interior point methods and metaheuristics for 0–1 programming

Our search deals with methods hybridizing interior point processes and metaheuristics for solving 0–1 linear programs. This paper shows how metaheuristics can take advantage of a sequence of interior points generated by an interior point method. After introducing our work field, we present our hybrid search which generates a diversified population. Next, we explain the whole method combining the solutions encountered in the previous phase through a path relinking template. Computational experiments are reported on 0–1 multiconstraint knapsack problems.     

Investigation of Ant System parameter interactions by using design of experiments for job-shop scheduling problems

In recent years, one of the most important and promising research fields has been metaheuristics to find optimal or near-optimal solutions for NP-hard combinatorial optimization problems. Improving the quality of the solution or the solution time is basic research area on metaheuristics. Modifications of the existing ones or creation of hybrid approaches are the focus of these efforts. Another area of improving the solution quality of metaheuristics is finding the optimal combination of algorithm control parameters. This is usually done by design of experiments or one-at-a-time approach in genetic algorithms, simulated annealing and similar metaheuristics. We observe that, in studies which use Ant Colonies Optimization (ACO) as an optimization technique; the levels of control parameters are determined by some non-systematic initial experiments and the interactions of the parameters are not studied yet.In this study, the parameters of Ant System have been investigated on different sized and randomly generated job-shop scheduling problems by using design of experiments. The effects and interactions of the parameters have been interpreted with the outputs of the experiments. Referring to the statistical analysis it is observed that none of the interactions between the Ant System parameters has a significant effect on makespan value. A specific fractional experimental design is suggested instead of the full factorial design. Depending on the findings from the benchmark problems it will be a reliable approach to use the suggested design for saving time and effort in experiments without sacrificing the solution quality.     

A parallel tabu search for solving the primal buffer allocation problem in serial production systems

This paper presents a novel parallel tabu search (PTS) algorithm equipped with a proper adaptive neighborhood generation mechanism to solve the primal buffer allocation problem, which consists of minimizing the total buffer capacity of a serial production system under a minimum throughput rate constraint. An evaluative method based on a specific algorithm has been implemented to simulate the system behavior. In order to validate the effectiveness of the proposed PTS a mixed integer linear programming-based simulation/optimization approach and several metaheuristics from the relevant literature have been implemented. Since most metaheuristics are sensitive to the parameter setting, a proper calibration analysis based on a non-parametric test has been performed. Then, a comprehensive comparison analysis, concerning with both quality of solutions and computational efficiency, has been carried out. Finally, through the numerical results obtained from PTS, a multi-factorial experimental analysis has been developed to analyze the influencing factors of the problem under investigation.     

A parameterized shared-memory scheme for parameterized metaheuristics

This paper presents a parameterized shared-memory scheme for parameterized metaheuristics. The use of a parameterized metaheuristic facilitates experimentation with different metaheuristics and hybridation/combinations to adapt them to the particular problem we are working with. Due to the large number of experiments necessary for the metaheuristic selection and tuning, parallelism should be used to reduce the execution time. To obtain parallel versions of the metaheuristics and to adapt them to the characteristics of the parallel system, a unified parameterized shared-memory scheme is developed. Given a particular computational system and fixed parameters for the sequential metaheuristic, the appropriate selection of parameters in the unified parallel scheme eases the development of parallel efficient metaheuristics.     

Optimization algorithms for large-scale real-world instances of the frequency assignment problem

Nowadays, mobile communications are experiencing a strong growth, being more and more indispensable. One of the key issues in the design of mobile networks is the frequency assignment problem (FAP). This problem is crucial at present and will remain important in the foreseeable future. Real-world instances of FAP typically involve very large networks, which can be handled only by heuristic methods. In the present work, we are interested in optimizing frequency assignments for problems described in a mathematical formalism that incorporates actual interference information, measured directly on the field, as is done in current GSM networks. To achieve this goal, a range of metaheuristics have been designed, adapted, and rigourously compared on two actual GSM networks modeled according to the latter formalism. To generate quickly and reliably high-quality solutions, all metaheuristics combine their global search capabilities with a local-search method specially tailored for this domain. The experiments and statistical tests show that in general, all metaheuristics are able to improve upon results published in previous studies, but two of the metaheuristics emerge as the best performers: a population-based algorithm (Scatter Search) and a trajectory based (1+1) Evolutionary Algorithm. Finally, the analysis of the frequency plans obtained offers insight about how the interference cost is reduced in the optimal plans.     

Solving the multidimensional knapsack problems with generalized upper bound constraints by the adaptive memory projection method

An adaptive memory projection (referred as AMP) method is developed for multidimensional knapsack problems (referred as the MKP) with generalized upper bound constraints. All the variables are divided into several generalized upper bound (referred as GUB) sets and at most one variable can be chosen from each of the GUB sets. The MKP with GUBs (referred as the GUBMKP) can be applied to many real-world problems, such as capital budgeting, resource allocation, cargo loading, and project selection. Due to the complexity of the GUBMKP, good metaheuristics are sought to tackle this problem.The AMP method keeps track of components of good solutions during the search and creates provisional solution by combining components of better solutions. The projection method, which can free the selected variables while fixing the others, is very useful for metaheuristics, especially when tackling large-scale combinatorial optimization. In this paper, the AMP method is implemented by iteratively using critical event tabu search as a search routine, and CPLEX in the referent optimization stage. Variables that are strongly determined, consistent, or attractive, are identified in the search process. Selected variables from this pool are fed into CPLEX as a small subproblem. In addition to the diversification effect within critical event tabu search, the pseudo-cut inequalities and an adjusted frequency penalty scalar are also applied to increase opportunities of exploring new regions.This study conducts a comprehensive sensitivity analysis on the parameters and strategies used in the proposed AMP method. The computational results show several variants of the AMP method outperforms the tight oscillation method in the literature of GUBMKP. On average, consistent variables tend to perform best as a pure strategy. A pure strategy equipped with local search can lead into even better results. Last but not least, testing different types of variables in the referent optimization stage before selecting just one of the pure strategies is found to be very helpful.     

Soft computing and cooperative strategies for optimization

In the last years, metaheuristics have emerged as powerful algorithmic approaches which have been applied with great success to difficult combinatorial optimization problems. However, this does not mean that metaheuristics can be applied blindly to any new problem.In this contribution we showed how the most basic ingredients of Soft Computing, namely fuzzy sets and fuzzy rules, are used in the context of a simple metaheuristic and a cooperative strategy based on it, to obtain successful results for the p-median problem.     

Achieving super-linear performance in parallel multi-objective evolutionary algorithms by means of cooperative coevolution

This article introduces three new multi-objective cooperative coevolutionary variants of three state-of-the-art multi-objective evolutionary algorithms, namely, Non-dominated Sorting Genetic Algorithm II (NSGA-II), Strength Pareto Evolutionary Algorithm 2 (SPEA2) and Multi-objective Cellular Genetic Algorithm (MOCell). In such a coevolutionary architecture, the population is split into several subpopulations or islands, each of them being in charge of optimizing a subset of the global solution by using the original multi-objective algorithm. Evaluation of complete solutions is achieved through cooperation, i.e., all subpopulations share a subset of their current partial solutions. Our purpose is to study how the performance of the cooperative coevolutionary multi-objective approaches can be drastically increased with respect to their corresponding original versions. This is specially interesting for solving complex problems involving a large number of variables, since the problem decomposition performed by the model at the island level allows for much faster executions (the number of variables to handle in every island is divided by the number of islands). We conduct a study on a real-world problem related to grid computing, the bi-objective robust scheduling problem of independent tasks. The goal in this problem is to minimize makespan (i.e., the time when the latest machine finishes its assigned tasks) and to maximize the robustness of the schedule (i.e., its tolerance to unexpected changes on the estimated time to complete the tasks). We propose a parallel, multithreaded implementation of the coevolutionary algorithms and we have analyzed the results obtained in terms of both the quality of the Pareto front approximations yielded by the techniques as well as the resulting speedups when running them on a multicore machine.     

An optimization algorithm inspired by social creativity systems

The need for efficient and effective optimization problem solving methods arouses nowadays the design and development of new heuristic algorithms. This paper present ideas that leads to a novel multiagent metaheuristic technique based on creative social systems suported on music composition concepts. This technique, called “Musical Composition Method” (MMC), which was proposed in Mora-Gutiérrez et?al. (Artif Intell Rev 2012) as well as a variant, are presented in this study. The performance of MMC is evaluated and analyzed over forty instances drawn from twenty-two benchmark global optimization problems. The solutions obtained by the MMC algorithm were compared with those of various versions of particle swarm optimizer and harmony search on the same problem set. The experimental results demonstrate that MMC significantly improves the global performances of the other tested metaheuristics on this set of multimodal functions.     

面向自主网格的多Agent动态协同图构造算法

网格环境的复杂性和动态性迫切需要自主计算技术的支持。在前期工作中给出地自主网格体系结构基础上,为解决任务执行过程中资源或服务失效情况下多Agent间协同的问题,提出了多Agent动态协同图的概念和任务偏序集驱动地多Agent动态协同图构造算法。图中的顶点是由Agent和自主网格服务组成的序偶,构造算法由任务偏序集到服务集的映射,逐层构造图中的顶点。该图通过Agent对本地服务的感知和Agent间的通信,达到任务执行过程中服务间自主协同的目标。模拟实验的结果验证了算法的正确性,表明算法的时间复杂度主要由任务哈斯图的层数决定,并且Agent的感知时间具有鲁棒性。     

Solving the feeder bus network design problem by genetic algorithms and ant colony optimization

This paper proposes the design and analysis of two metaheuristics, genetic algorithms and ant colony optimization, for solving the feeder bus network design problem. A study of how these proposed heuristics perform is carried out on several randomly generated test problems to evaluate their computational efficiency and the quality of solutions obtained by them. The results are also compared to those published in the literature. Computational experiments have shown that both heuristics are comparable to the state-of-the-art algorithms such as simulated annealing and tabu search.     

A comprehensive survey on optimizing deep learning models by metaheuristics

Deep neural networks (DNNs), which are extensions of artificial neural networks, can learn higher levels of feature hierarchy established by lower level features by transforming the raw feature space to another complex feature space. Although deep networks are successful in a wide range of problems in different fields, there are some issues affecting their overall performance such as selecting appropriate values for model parameters, deciding the optimal architecture and feature representation and determining optimal weight and bias values. Recently, metaheuristic algorithms have been proposed to automate these tasks. This survey gives brief information about common basic DNN architectures including convolutional neural networks, unsupervised pre-trained models, recurrent neural networks and recursive neural networks. We formulate the optimization problems in DNN design such as architecture optimization, hyper-parameter optimization, training and feature representation level optimization. The encoding schemes used in metaheuristics to represent the network architectures are categorized. The evolutionary and selection operators, and also speed-up methods are summarized, and the main approaches to validate the results of networks designed by metaheuristics are provided. Moreover, we group the studies on the metaheuristics for deep neural networks based on the problem type considered and present the datasets mostly used in the studies for the readers. We discuss about the pros and cons of utilizing metaheuristics in deep learning field and give some future directions for connecting the metaheuristics and deep learning. To the best of our knowledge, this is the most comprehensive survey about metaheuristics used in deep learning field.

     

A new heuristic approach for non-convex optimization problems

In this work a new optimization method, called the heuristic Kalman algorithm (HKA), is presented. This new algorithm is proposed as an alternative approach for solving continuous, non-convex optimization problems. The principle of HKA is to explicitly consider the optimization problem as a measurement process designed to give an estimate of the optimum. A specific procedure, based on the Kalman estimator, was developed to improve the quality of the estimate obtained through the measurement process. The main advantage of HKA, compared to other metaheuristics, lies in the small number of parameters that need to be set by the user. Further, it is shown that HKA converges almost surely to a near-optimal solution. The efficiency of HKA was evaluated in detail using several non-convex test problems, both in the unconstrained and constrained cases. The results were then compared to those obtained via other metaheuristics. The numerical experiments show that HKA is a promising approach for solving non-convex optimization problems, particularly in terms of computation time and success ratio.     

协同设计环境下的版本管理模型

计算机支持协同设计（computer supported cooperative design）是计算机支持协同工作CSCW的一个应用领域,也是CAD技术在未来发展的趋势。由于协同设计中多个人员共同参与设计,必然引进约束来保证协同。需要对设计对象和约束进行协调管理。协同设计中设计对象存在着多个版本。如何保证约束和设计对象版本间的一致对于协同设计的顺利进行具有重要意义。针对这些特点,作者提出了一种基于约束的层次化的管理协同设计版本的模型。     

一种引入局部交互的群体协作行为协同进化机制

现有协同进化模型在求解子系统间存在相互关联作用的问题时存在不足,使其难以高效产生群体复杂适应性协作行为．针对这一问题,依据系统论和非线性科学理论,构造了一种引入局部交互的群体复杂协作行为协同进化机制．该机制通过局部交互作用探测局部启发信息,并结合全局启发信息共同引导协同进化过程,从而使求解过程朝着正确的方向进化．算法分析及多机器人协作推箱实验表明,该机制模型及其算法有效地克服了现有协同进化模型的不足和局限,能使复杂关联的群体协作行为高效地朝着全局最优协作方向演化．     

二进制中心引力优化算法及其在非线性0-1规划中的应用

中心引力优化算法是一种新型智能优化算法,源于万有引力定律的模拟。该算法和现有其他智能优化算法最显著的区别是其完全确定性的优化特征,不含有任何随机因素。为求解非线性0-1规划问题,本文提出了一种二进制中心引力优化算法。根据引力计算加速度,利用加速度更新位置,采用转换函数实现连续的位置变量到离散的0-1变量的变换。采用典型的非线性0-1规划测试问题进行数值实验,并将算法与二进制粒子群优化算法和二进制引力搜索算法进行比较。实验结果表明在解的稳定性和计算精度两个方面本文给出的算法具有显著优势,为非线性0-1规划问题的求解提供了新方法。     

Exploiting multilevel parallelism on a many-core system for the application of hyperheuristics to a molecular docking problem

The solution of Protein–Ligand Docking Problems can be approached through metaheuristics, and satisfactory metaheuristics can be obtained with hyperheuristics searching in the space of metaheuristics implemented inside a parameterized schema. These hyperheuristics apply several metaheuristics, resulting in high computational costs. To reduce execution times, a shared-memory schema of hyperheuristics is used with four levels of parallelism, two for the hyperheuristic and two for the metaheuristics. The parallel schema is executed in a many-core system in “native mode,” and the four-level parallelism allows us to take full advantage of the massive parallelism offered by this architecture and obtain satisfactory fitness and an important reduction in the execution time.