On-the-fly calibrating strategies for evolutionary algorithms

The issue of controlling values of various parameters of an evolutionary algorithm is one of the most important and interesting areas of research in evolutionary computation. In this paper we propose two new parameter control strategies for evolutionary algorithms based on the ideas of reinforcement learning. These strategies provide efficient and low-cost adaptive techniques for parameter control and they preserve the original design of the evolutionary algorithm, as they can be included without changing either the structure of the algorithm nor its operators design.     

Predictive reference shaping for constrained robotic systems using evolutionary algorithms

This paper proposes a method for reducing the trajectory tracking errors of robotic systems in presence of input saturation and state constraints. Basing on a finite horizon prediction of the future evolution of the robot dynamics, the proposed device online preshapes the reference trajectory, minimizing a multi-objective cost function. The shaped reference is updated at discrete time intervals and is generated taking into account the full nonlinear robot dynamics, input and state constraints. A specialized Evolutionary Algorithm is employed as search tool for the online computation of a sub-optimal reference trajectory in the discretized space of the control alternatives. The effectiveness of the proposed method and the online computational burden are analyzed numerically in two significant robotic control problems; furthermore a comparison of the performance provided by this method and an iterative gradient-based algorithms are discussed.     

Fuzzy assembly line balancing using genetic algorithms

In this paper, we implement genetic algorithms to synthesis fuzzy assembly line balancing problem which is well-known as a NP-hard problem. The genetic operators concerned with the feasibility of chromosomes will be discussed, and its performance will be shown with a numerical example.     

Simplifying the automatic design of a fuzzy logic controller using evolutionary programming

With the availability of a wide range of Evolutionary Algorithms such as Genetic Algorithms, Evolutionary Programming, Evolutionary Strategies and Differential Evolution, every conceivable aspect of the design of a fuzzy logic controller has been optimized and automated. Although there is no doubt that these automated techniques can produce an optimal fuzzy logic controller, the structure of such a controller is often obscure and in many cases these optimizations are simply not needed. We believe that the automatic design of a fuzzy logic controller can be simplified by using a generic rule base such as the MacVicar-Whelan rule base and using an evolutionary algorithm to optimize only the membership functions of the fuzzy sets. Furthermore, by restricting the overlapping of fuzzy sets, using triangular membership functions and singletons, and reducing the number of parameters to represent the membership functions, the design can be further simplified. This paper describes this method of simplifying the design and some experiments performed to ascertain its validity.     

An experimental study of adaptive control for evolutionary algorithms

In this paper, we investigate how adaptive operator selection techniques are able to efficiently manage the balance between exploration and exploitation in an evolutionary algorithm, when solving combinatorial optimization problems. We introduce new high level reactive search strategies based on a generic algorithm's controller that is able to schedule the basic variation operators of the evolutionary algorithm, according to the observed state of the search. Our experiments on SAT instances show that reactive search strategies improve the performance of the solving algorithm.     

Studying the effect of population size in distributed evolutionary algorithms on heterogeneous clusters

Distributed Evolutionary Algorithms are traditionally executed on homogeneous dedicated clusters, despite most scientists have access mainly to networks of heterogeneous nodes (e.g., desktop PCs in a lab). Fitting this kind of algorithms to these environments, so that they can take advantage of their heterogeneity to save running time, is still an open problem. The different computational power of the nodes affects the performance of the algorithm, and tuning or fitting it to each node properly could reduce execution time.Since the distributed Evolutionary Algorithms include a whole range of parameters that influence the performance, this paper proposes a study on the population size. This parameter is one of the most important, since it has a direct relationship with the number of iterations needed to find the solution, as it affects the exploration factor of the algorithm. The aim of this paper consists in validating the following hypothesis: fitting the sub-population size to the computational power of the heterogeneous cluster node can lead to an improvement in running time with respect to the use of the same population size in every node.Two parameter size schemes have been tested, an offline and an online parameter setting, and three problems with different characteristics and computational demands have been used.Results show that setting the population size according to the computational power of each node in the heterogeneous cluster improves the time required to obtain the optimal solution. Meanwhile, the same set of different size values could not improve the running time to reach the optimum in a homogeneous cluster with respect to the same size in all nodes, indicating that the improvement is due to the interaction of the different hardware resources with the algorithm. In addition, a study on the influence of the different population sizes on each stage of the algorithm is presented. This opens a new research line on the fitting (offline or online) of parameters of the distributed Evolutionary Algorithms to the computational power of the devices.     

A variable reduction strategy for evolutionary algorithms handling equality constraints

Efficient constraint handling techniques are of great significance when Evolutionary Algorithms (EAs) are applied to constrained optimization problems (COPs). Generally, when use EAs to deal with COPs, equality constraints are much harder to satisfy, compared with inequality constraints. In this study, we propose a strategy named equality constraint and variable reduction strategy (ECVRS) to reduce equality constraints as well as variables of COPs. Since equality constraints are always expressed by equations, ECVRS makes use of the variable relationships implied in such equality constraint equations. The essence of ECVRS is it makes some variables of a COP considered be represented and calculated by some other variables, thereby shrinking the search space and leading to efficiency improvement for EAs. Meanwhile, ECVRS eliminates the involved equality constraints that providing variable relationships, thus improves the feasibility of obtained solutions. ECVRS is tested on many benchmark problems. Computational results and comparative studies verify the effectiveness of the proposed ECVRS.     

PLASMA自适应调优与性能优化的设计与实现

PLASMA是一个高效的线性代数软件包,其数据分布结合分堆、细粒度并行以及乱序执行机制等大大提高了程序的性能。但PLASMA仍然存在一些问题,比如分块大小对程序性能的影响非常大,以及产生了大量的数据拷贝等。通过对比传统的LAPACK和PLASMA的实现机制,分析了PLASMA中存在的优势和不足,介绍了两种弥补PLASMA自身不足的方法。针对PLASMA的架构,经过大量的测试与分析,提出了边缘矩阵的概念并分析了其对性能的影响,据此提出了一种自适应调优的方法。并通过数据拷贝与计算并行的运行方式,进一步提高了PLASMA性能,最后通过大量的测试验证了该优化方法的效果。     

Self-adaptivity for grid applications. An Efficient Resources Selection model based on evolutionary computation algorithms

Over the last few years, the adaptation ability has become an essential characteristic for grid applications due to the fact that it allows applications to face the dynamic and changing nature of grid systems. This adaptive capability is applied within different grid processes such as resource monitoring, resource discovery, or resource selection. In this regard, the present approach provides a self-adaptive ability to grid applications, focusing on enhancing the resources selection process. This contribution proposes an Efficient Resources Selection model to determine the resources that best fit the application requirements. Hence, the model guides applications during their execution without modifying or controlling grid resources. Within the evaluation phase, the experiments were carried out in a real European grid infrastructure. Finally, the results show that not only a self-adaptive ability is provided by the model but also a reduction in the applications’ execution time and an improvement in the successfully completed tasks rate are accomplished.     

Hybrid evolutionary method for obstacle location-allocation

In this paper we propose a hybrid evolutionary method for Obstacle Location-allocation problem. This problem can be described as a tri-level mixed integer programming problem. Since this problem is very complex and with many local solutions, no direct method is effective to solve it. Heuristic methods were proposed to it, but optimality is not guaranteed yet. Our hybrid evolutionary method adopts the main structure of Genetic Algorithms (GA) absorbing ideas from Evolutionary Strategy (ES) and combines with some traditional optimization techniques. In this way we can pursue global optimization maintaining a good efficiency of our method. A case study shows the effectiveness of this method.     

Design of local fuzzy models using evolutionary algorithms

The application of local fuzzy models to determine the remaining life of a unit in a fleet of vehicles is described. Instead of developing individual models based on the track history of each unit or developing a global model based on the collective track history of the fleet, local fuzzy models are used based on clusters of peers—similar units with comparable utilization and performance characteristics. A local fuzzy performance model is created for each cluster of peers. This is combined with an evolutionary framework to maintain the models. A process has been defined to generate a collection of competing models, evaluate their performance in light of the currently available data, refine the best models using evolutionary search, and select the best one after a finite number of iterations. This process is repeated periodically to automatically update and improve the overall model. To illustrate this methodology an asset selection problem has been identified: given a fleet of industrial vehicles (diesel electric locomotives), select the best subset for mission-critical utilization. To this end, the remaining life of each unit in the fleet is predicted. The fleet is then sorted using this prediction and the highest ranked units are selected. A series of experiments using data from locomotive operations was conducted and the results from an initial validation exercise are presented. The approach of constructing local predictive models using fuzzy similarity with neighboring points along appropriate dimensions is not specific to any asset type and may be applied to any problem where the premise of similarity along chosen attribute dimensions implies similarity in predicted future behavior.     

Hybrid evolutionary method for capacitated location-allocation problem

Location-allocation model is widely applied for facility location design in practice. In this paper, we discuss an extension of location-allocation model which has capacity constraints and propose a hybrid evolutionary method to solve it which absorbs ideas from both genetic algorithms (GAs) and evolutionary strategy (ES) as well as combined with efficient traditional optimization techniques. It is shown that the proposed method is effective in finding global or near global solutions by numerical simulations.     

A new hybrid method using evolutionary algorithms to train Fuzzy Cognitive Maps

A novel hybrid method based on evolutionary computation techniques is presented in this paper for training Fuzzy Cognitive Maps. Fuzzy Cognitive Maps is a soft computing technique for modeling complex systems, which combines the synergistic theories of neural networks and fuzzy logic. The methodology of developing Fuzzy Cognitive Maps relies on human expert experience and knowledge, but still exhibits weaknesses in utilization of learning methods and algorithmic background. For this purpose, we investigate a coupling of differential evolution algorithm and unsupervised Hebbian learning algorithm, using both the global search capabilities of Evolutionary strategies and the effectiveness of the nonlinear Hebbian learning rule. The use of differential evolution algorithm is related to the concept of evolution of a number of individuals from generation to generation and that of nonlinear Hebbian rule to the concept of adaptation to the environment by learning. The hybrid algorithm is introduced, presented and applied successfully in real-world problems, from chemical industry and medicine. Experimental results suggest that the hybrid strategy is capable to train FCM effectively leading the system to desired states and determining an appropriate weight matrix for each specific problem.     

On the equivalences and differences of evolutionary algorithms

Evolutionary algorithms (EAs) are fast and robust computation methods for global optimization, and have been widely used in many real-world applications. We first conceptually discuss the equivalences of various popular EAs including genetic algorithm (GA), biogeography-based optimization (BBO), differential evolution (DE), evolution strategy (ES) and particle swarm optimization (PSO). We find that the basic versions of BBO, DE, ES and PSO are equal to the GA with global uniform recombination (GA/GUR) under certain conditions. Then we discuss their differences based on biological motivations and implementation details, and point out that their distinctions enhance the diversity of EA research and applications. To further study the characteristics of various EAs, we compare the basic versions and advanced versions of GA, BBO, DE, ES and PSO to explore their optimization ability on a set of real-world continuous optimization problems. Empirical results show that among the basic versions of the algorithms, BBO performs best on the benchmarks that we studied. Among the advanced versions of the algorithms, DE and ES perform best on the benchmarks that we studied. However, our main conclusion is that the conceptual equivalence of the algorithms is supported by the fact that algorithmic modifications result in very different performance levels.     

Accelerated diffusion algorithms for dynamic load balancing

In this paper we consider the application of accelerated techniques in order to increase the rate of convergence of the diffusive iterative load balancing algorithms. In particular, we compare the application of Semi-Iterative, Second Degree and Variable Extrapolation techniques on the basic diffusion method for various types of network graphs.     

A parameter control method of evolutionary algorithms using exploration and exploitation measures with a practical application for fitting Sovova's mass transfer model

Exploration and exploitation are omnipresent terms in evolutionary computation community that have been broadly utilized to explain how evolutionary algorithms perform search. However, only recently exploration and exploitation measures were presented in a quantitative way enabling to measure amounts of exploration and exploitation. To move a step further, this paper introduces a parameter control approach that utilizes such measures as feedback to adaptively control evolution processes. The paper shows that with new exploration and exploitation measures, the evolution process generates relatively well results in terms of fitness and/or convergence rate when applying to a practical chemical engineering problem of fitting Sovova's model. We also conducted an objective statistical analysis using Bonferroni–Dunn test and sensitivity analysis on the experimental results. The statistical analysis results again proved that the parameter control strategy using exploration and exploitation measures is competitive to the other approaches presented in the paper. The sensitivity analysis results also showed that different initial values may affect output in different magnitude.     

A Flexible Evolutionary Agent: cooperation and competition among real-coded evolutionary operators

Since it has currently became essential to design more efficient and robust alternative techniques to solve hard optimisation problems in industry or science, and of easy use for practitioners, here a new way of developing simple Artificial Intelligence based Evolutionary Algorithms will be introduced. Our evolutionary computational implementation is a new idea in optimisation. Any evolutionary operators and their associated parameters from well-established evolutionary methods can be considered in such a way that the entire algorithm or intelligent agent-based software performs with very high efficiency without a prior need to investigate which method will be the best for a given optimisation problem.The implementation presented, named Flexible Evolution (FE), has capacity to adapt the operators, the parameters and the algorithm to the circumstances faced at each step of every optimisation run and is able to take into account lessons learned by different research works in the adaptation of operators and parameters. The FE uses Artificial Intelligence concepts to manage internal procedures to adopt decisions and correct the wrong ones. Our aim in this paper will be to give the keys to design these types of procedures, and more specifically, to find the way of achieving an optimum performance of the operators involved in the search, in our case by means of a function included in our algorithm called Sampling Engine. An early implementation has been already developed and tested in our previous works [66–68], so in this paper, new results of a second software implementation are presented comparing the results with those obtained by other methods, using well-known hard test functions.     

Observations in using parallel and sequential evolutionary algorithms for automatic software testing

In this paper we analyze the application of parallel and sequential evolutionary algorithms (EAs) to the automatic test data generation problem. The problem consists of automatically creating a set of input data to test a program. This is a fundamental step in software development and a time consuming task in existing software companies. Canonical sequential EAs have been used in the past for this task. We explore here the use of parallel EAs. Evidence of greater efficiency, larger diversity maintenance, additional availability of memory/CPU, and multi-solution capabilities of the parallel approach, reinforce the importance of the advances in research with these algorithms. We describe in this work how canonical genetic algorithms (GAs) and evolutionary strategies (ESs) can help in software testing, and what the advantages are (if any) of using decentralized populations in these techniques. In addition, we study the influence of some parameters of the proposed test data generator in the results. For the experiments we use a large benchmark composed of twelve programs that includes fundamental algorithms in computer science.     

A new evolutionary search strategy for global optimization of high-dimensional problems

Global optimization of high-dimensional problems in practical applications remains a major challenge to the research community of evolutionary computation. The weakness of randomization-based evolutionary algorithms in searching high-dimensional spaces is demonstrated in this paper. A new strategy, SP-UCI is developed to treat complexity caused by high dimensionalities. This strategy features a slope-based searching kernel and a scheme of maintaining the particle population’s capability of searching over the full search space. Examinations of this strategy on a suite of sophisticated composition benchmark functions demonstrate that SP-UCI surpasses two popular algorithms, particle swarm optimizer (PSO) and differential evolution (DE), on high-dimensional problems. Experimental results also corroborate the argument that, in high-dimensional optimization, only problems with well-formative fitness landscapes are solvable, and slope-based schemes are preferable to randomization-based ones.     

Ranking and comparing evolutionary algorithms with Hellinger-TOPSIS

When multiple algorithms are applied to multiple benchmarks as it is common in evolutionary computation, a typical issue rises, how can we rank the algorithms? It is a common practice in evolutionary computation to execute the algorithms several times and then the mean value and the standard deviation are calculated. In order to compare the algorithms performance it is very common to use statistical hypothesis tests. In this paper, we propose a novel alternative method based on the Technique for Order Preference by Similarity to Ideal Solution (TOPSIS) to support the performance comparisons. In this case, the alternatives are the algorithms and the criteria are the benchmarks. Since the standard TOPSIS is not able to handle the stochastic nature of evolutionary algorithms, we apply the Hellinger-TOPSIS, which uses the Hellinger distance, for algorithm comparisons. Case studies are used to illustrate the method for evolutionary algorithms but the approach is general. The simulation results show the feasibility of the Hellinger-TOPSIS to find out the ranking of algorithms under evaluation.