Showing papers on "Evolutionary programming published in 2011"

Surrogate-assisted evolutionary computation: Recent advances and future challenges

Abstract: Surrogate-assisted, or meta-model based evolutionary computation uses efficient computational models, often known as surrogates or meta-models, for approximating the fitness function in evolutionary algorithms. Research on surrogate-assisted evolutionary computation began over a decade ago and has received considerably increasing interest in recent years. Very interestingly, surrogate-assisted evolutionary computation has found successful applications not only in solving computationally expensive single- or multi-objective optimization problems, but also in addressing dynamic optimization problems, constrained optimization problems and multi-modal optimization problems. This paper provides a concise overview of the history and recent developments in surrogate-assisted evolutionary computation and suggests a few future trends in this research area.

Quantum-inspired evolutionary algorithms: a survey and empirical study

Abstract: Quantum-inspired evolutionary algorithms, one of the three main research areas related to the complex interaction between quantum computing and evolutionary algorithms, are receiving renewed attention. A quantum-inspired evolutionary algorithm is a new evolutionary algorithm for a classical computer rather than for quantum mechanical hardware. This paper provides a unified framework and a comprehensive survey of recent work in this rapidly growing field. After introducing of the main concepts behind quantum-inspired evolutionary algorithms, we present the key ideas related to the multitude of quantum-inspired evolutionary algorithms, sketch the differences between them, survey theoretical developments and applications that range from combinatorial optimizations to numerical optimizations, and compare the advantages and limitations of these various methods. Finally, a small comparative study is conducted to evaluate the performances of different types of quantum-inspired evolutionary algorithms and conclusions are drawn about some of the most promising future research developments in this area.

Short term wind speed prediction based on evolutionary support vector regression algorithms

Abstract: Hyper-parameters estimation in regression Support Vector Machines (SVMr) is one of the main problems in the application of this type of algorithms to learning problems This is a hot topic in which very recent approaches have shown very good results in different applications in fields such as bio-medicine, manufacturing, control, etc Different evolutionary approaches have been tested to be hybridized with SVMr, though the most used are evolutionary approaches for continuous problems, such as evolutionary strategies or particle swarm optimization algorithms In this paper we discuss the application of two different evolutionary computation techniques to tackle the hyper-parameters estimation problem in SVMrs Specifically we test an Evolutionary Programming algorithm (EP) and a Particle Swarm Optimization approach (PSO) We focus the paper on the discussion of the application of the complete evolutionary-SVMr algorithm to a real problem of wind speed prediction in wind turbines of a Spanish wind farm

The mean–variance cardinality constrained portfolio optimization problem: An experimental evaluation of five multiobjective evolutionary algorithms

Abstract: This paper compares the effectiveness of five state-of-the-art multiobjective evolutionary algorithms (MOEAs) together with a steady state evolutionary algorithm on the mean–variance cardinality constrained portfolio optimization problem (MVCCPO). The main computational challenges of the model are due to the presence of a nonlinear objective function and the discrete constraints. The MOEAs considered are the Niched Pareto genetic algorithm 2 (NPGA2), non-dominated sorting genetic algorithm II (NSGA-II), Pareto envelope-based selection algorithm (PESA), strength Pareto evolutionary algorithm 2 (SPEA2), and e-multiobjective evolutionary algorithm (e-MOEA). The computational comparison was performed using formal metrics proposed by the evolutionary multiobjective optimization community on publicly available data sets which contain up to 2196 assets.

Efficiency issues of evolutionary k-means

Abstract: One of the top ten most influential data mining algorithms, k-means, is known for being simple and scalable. However, it is sensitive to initialization of prototypes and requires that the number of clusters be specified in advance. This paper shows that evolutionary techniques conceived to guide the application of k-means can be more computationally efficient than systematic (i.e., repetitive) approaches that try to get around the above-mentioned drawbacks by repeatedly running the algorithm from different configurations for the number of clusters and initial positions of prototypes. To do so, a modified version of a (k-means based) fast evolutionary algorithm for clustering is employed. Theoretical complexity analyses for the systematic and evolutionary algorithms under interest are provided. Computational experiments and statistical analyses of the results are presented for artificial and text mining data sets.

A multilevel evolutionary algorithm for optimizing numerical functions

Abstract: Article history: Received 15 April 2010 Received in revised form 15 May 2010 Accepted 16 May 2010 Available online 17 May 2010 This is a study on the effects of multilevel selection (MLS) theory in optimizing numerical functions. Based on this theory, a Multilevel Evolutionary Optimization algorithm (MLEO) is presented. In MLEO, a species is subdivided in cooperative populations and then each population is subdivided in groups, and evolution occurs at two levels so called individual and group levels. A fast population dynamics occurs at individual level. At this level, selection occurs among individuals of the same group. The popular genetic operators such as mutation and crossover are applied within groups. A slow population dynamics occurs at group level. At this level, selection happens among groups of a population. The group level operators such as regrouping, migration, and extinction-colonization are applied among groups. In regrouping process, all the groups are mixed together and then new groups are formed. The migration process encourages an individual to leave its own group and move to one of its neighbour groups. In extinction-colonization process, a group is selected as extinct, and replaced by offspring of a colonist group. In order to evaluate MLEO, the proposed algorithms were used for optimizing a set of well known numerical functions. The preliminary results indicate that the MLEO theory has positive effect on the evolutionary process and provide an efficient way for numerical optimization. © 2011 Growing Science Ltd. All rights reserved.

Evolving behaviour trees for the Mario AI competition using grammatical evolution

Abstract: This paper investigates the applicability of Genetic Programming type systems to dynamic game environments. Grammatical Evolution was used to evolved Behaviour Trees, in order to create controllers for the Mario AI Benchmark. The results obtained reinforce the applicability of evolutionary programming systems to the development of artificial intelligence in games, and in dynamic systems in general, illustrating their viability as an alternative to more standard AI techniques.

Approximation-guided evolutionary multi-objective optimization

Abstract: Multi-objective optimization problems arise frequently in applications but can often only be solved approximately by heuristic approaches. Evolutionary algorithms have been widely used to tackle multi-objective problems. These algorithms use different measures to ensure diversity in the objective space but are not guided by a formal notion of approximation. We present a new framework of an evolutionary algorithm for multi-objective optimization that allows to work with a formal notion of approximation. Our experimental results show that our approach outperforms state-of-the-art evolutionary algorithms in terms of the quality of the approximation that is obtained in particular for problems with many objectives.

Evolutionary Sets Of Safe Ship Trajectories: A New Approach To Collision Avoidance

Abstract: The paper introduces a new method of solving multi-ship encounter situations for both open waters and restricted water regions. The method, called evolutionary sets of safe trajectories, combines some of the assumptions of game theory with evolutionary programming and aims to find optimal sets of safe trajectories of all ships involved in an encounter situation. In a two-ship encounter situation it enables the operator of an onboard collision-avoidance system to predict the most probable behaviour of a target and to plan the own manoeuvres in advance. In a multi-ship encounter the method may be used to help an operator of a VTS system to coordinate the manoeuvres of all ships. The paper contains a detailed description of collision-avoidance operators used by the evolutionary method and simulation examples of the method's results for digital maps.

Adaptive Evolutionary Artificial Neural Networks for Pattern Classification

Abstract: This paper presents a new evolutionary approach called the hybrid evolutionary artificial neural network (HEANN) for simultaneously evolving an artificial neural networks (ANNs) topology and weights Evolutionary algorithms (EAs) with strong global search capabilities are likely to provide the most promising region However, they are less efficient in fine-tuning the search space locally HEANN emphasizes the balancing of the global search and local search for the evolutionary process by adapting the mutation probability and the step size of the weight perturbation This is distinguishable from most previous studies that incorporate EA to search for network topology and gradient learning for weight updating Four benchmark functions were used to test the evolutionary framework of HEANN In addition, HEANN was tested on seven classification benchmark problems from the UCI machine learning repository Experimental results show the superior performance of HEANN in fine-tuning the network complexity within a small number of generations while preserving the generalization capability compared with other algorithms

Short-term combined economic emission scheduling of hydrothermal systems with cascaded reservoirs using particle swarm optimization technique

Abstract: This paper develops an efficient and reliable particle swarm optimization (PSO) based algorithm for solving combined economic emission scheduling of hydrothermal systems with cascaded reservoirs. A multi-chain cascaded hydrothermal system with non-linear relationship between water discharge rate, power generation and net head is considered in this paper. The water transport delay between connected reservoirs is also considered. The problem is formulated considering both cost and emission as competing objectives. Combined economic emission scheduling (CEES) is a bi-objective problem. A price penalty factor approach is utilized here to convert this bi-objective CEES problem into a single objective one. The effect of valve-point loading is also taken into account in the present problem formulation. The feasibility of the proposed method is demonstrated on a sample test system consisting of four cascaded hydro units and three thermal units. The results of the proposed technique based on PSO are compared with other evolutionary programming method. It is found that the results obtained by the proposed technique are superior in terms of fuel cost, emission output etc. It is also observed that the computation time is considerably reduced by the proposed technique based on PSO.

Software Engineering Meets Evolutionary Computation

Abstract: The concept of evolutionary computation has affected virtually every area of software design, not merely as a metaphor, but as a realistic algorithm for exploration, insight, and improvement.

Quasi-oppositional Biogeography-based Optimization for Multi-objective Optimal Power Flow

Abstract: This article develops an efficient and reliable evolutionary programming algorithm, namely quasi-oppositional biogeography-based optimization, for solving optimal power flow problems. To improve the simulation results as well as the speed of convergence, opposition-based learning is incorporated in the original biogeography-based optimization algorithm. In order to investigate the performance, the proposed scheme is applied on optimal power flow problems of standard 26-bus, IEEE 118-bus, and IEEE 300-bus systems; and comparisons among mixed-integer particle swarm optimization, evolutionary programming, the genetic algorithm, original biogeography-based optimization, and quasi-oppositional biogeography-based optimization are presented. The results show that the new quasi-oppositional biogeography-based optimization algorithm outperforms the other techniques in terms of convergence speed and global search ability.

Adaptive Impedance Control for Upper-Limb Rehabilitation Robot Using Evolutionary Dynamic Recurrent Fuzzy Neural Network

Abstract: Control system implementation is one of the major difficulties in rehabilitation robot design. A newly developed adaptive impedance controller based on evolutionary dynamic fuzzy neural network (EDRFNN) is presented, where the desired impedance between robot and impaired limb can be regulated in real time according to the impaired limb's physical recovery condition. Firstly, the impaired limb's damping and stiffness parameters for evaluating its physical recovery condition are online estimated by using a slide average least squares (SALS)identification algorithm. Then, hybrid learning algorithms for EDRFNN impedance controller are proposed, which comprise genetic algorithm (GA), hybrid evolutionary programming (HEP) and dynamic back-propagation (BP) learning algorithm. GA and HEP are used to off-line optimize DRFNN parameters so as to get suboptimal impedance control parameters. Dynamic BP learning algorithm is further online fine-tuned based on the error gradient descent method. Moreover, the convergence of a closed loop system is proven using the discrete-type Lyapunov function to guarantee the global convergence of tracking error. Finally, simulation results show that the proposed controller provides good dynamic control performance and robustness with regard to the change of the impaired limb's physical condition.

Analytical Programming - a Novel Approach for Evolutionary Synthesis of Symbolic Structures

Abstract: This chapter discusses an alternative approach for symbolic structures and solutions synthesis and demonstrates a comparison with other methods, for example Genetic Programming (GP) or Grammatical Evolution (GE). Generally, there are two well known methods, which can be used for symbolic structures synthesis by means of computers. The first one is called GP and the other is GE. Another interesting research was carried out by Artificial Immune Systems (AIS) or/and systems, which do not use tree structures like linear GP and other similar algorithm like Multi Expression Programming (MEP), etc. In this chapter, a different method called Analytic Programming (AP), is presented. AP is a grammar free algorithmic superstructure, which can be used by any programming language and also by any arbitrary Evolutionary Algorithm (EA) or another class of numerical optimization method. This chapter describes not only theoretical principles of AP, but also its comparative study with selected well known case examples from GP as well as applications on synthesis of: controller, systems of deterministic chaos, electronics circuits, etc. For simulation purposes, AP has been co-joined with EA’s like Differential Evolution (DE), Self-Organising Migrating Algorithm (SOMA), Genetic Algorithms (GA) and Simulated Annealing (SA). All case studies has been carefully prepared and repeated in order to get valid statistical data for proper conclusions. The term symbolic regression represents a process during which measured data sets are fitted, thereby a corresponding mathematical formula is obtained in an analytical way. An output

Survival of dominated strategies under evolutionary dynamics

Abstract: We show that any evolutionary dynamic that satisfies three mild requirements— continuity, positive correlation, and innovation—does not eliminate strictly dominated strategies in all games. Likewise, we demonstrate that existing elimination results for evolutionary dynamics are not robust to small changes in the specifications of the dynamics.

Evolutionary Games on Star Graphs Under Various Updating Rules

Abstract: Evolutionary game dynamics have been traditionally studied in well-mixed populations where each individual is equally likely to interact with every other individual. Recent studies have shown that the outcome of the evolutionary process might be significantly affected if the population has a non-homogeneous structure. In this paper we study analytically an evolutionary game between two strategies interacting on an extreme heterogeneous graph, the star graph. We find explicit expressions for the fixation probability of mutants, and the time to absorption (elimination or fixation of mutants) and fixation (absorption conditional on fixation occurring). We investigate the evolutionary process considering four important update rules. For each of the update rules, we find appropriate conditions under which one strategy is favoured over the other. The process is considered in four different scenarios: the fixed fitness case, the Hawk–Dove game, the Prisoner’s dilemma and a coordination game. It is shown that in contrast with homogeneous populations, the choice of the update rule might be crucial for the evolution of a non-homogeneous population.

Neutrality in evolutionary algorithms... what do we know

Abstract: Over the last years, the effects of neutrality have attracted the attention of many researchers in the Evolutionary Algorithms (EAs) community. A mutation from one gene to another is considered as neutral if this modification does not affect the phenotype. This article provides a general overview on the work carried out on neutrality in EAs. Using as a framework the origin of neutrality and its study in different paradigms of EAs (e.g., Genetic Algorithms, Genetic Programming), we discuss the most significant works and findings on this topic. This work points towards open issues, which we belive the community needs to address.

Evolutionary dynamics in a simple model of self-assembly.

Abstract: We investigate the evolutionary dynamics of an idealized model for the robust self-assembly of two-dimensional structures called polyominoes The model includes rules that encode interactions between sets of square tiles that drive the self-assembly process The relationship between the model's rule set and its resulting self-assembled structure can be viewed as a genotype-phenotype map and incorporated into a genetic algorithm The rule sets evolve under selection for specified target structures The corresponding complex fitness landscape generates rich evolutionary dynamics as a function of parameters such as the population size, search space size, mutation rate, and method of recombination Furthermore, these systems are simple enough that in some cases the associated model genome space can be completely characterized, shedding light on how the evolutionary dynamics depends on the detailed structure of the fitness landscape Finally, we apply the model to study the emergence of the preference for dihedral over cyclic symmetry observed for homomeric protein tetramers

Do Evolutionary Algorithm Dynamics Create Complex Network Structures

Abstract: This paper presents a novel method for visualizing the dynamics of evolutionary algorithms in the form of complex networks. We discuss the analogy between individuals in populations in an arbitrary evolutionary algorithm and vertices of a complex network, as well as between edges in a complex network and communication between individuals in a population. The possibility of visualizing the dynamics of a complex network using the coupled map lattices method and control by means of chaos control techniques are also discussed.

A New Evolutionary Algorithm for a Class of Nonlinear Bilevel Programming Problems and Its Global Convergence

Abstract: When the leader's objective function of a nonlinear bilevel programming problem is nondifferentiable and the follower's problem of it is nonconvex, the existing algorithms cannot solve the problem. In this paper, a new effective evolutionary algorithm is proposed for this class of nonlinear bilevel programming problems. First, based on the leader's objective function, a new fitness function is proposed that can be easily used to evaluate the quality of different types of potential solutions. Then, based on Latin squares, an efficient crossover operator is constructed that has the ability of local search. Furthermore, a new mutation operator is designed by using some good search directions so that the offspring can approach a global optimal solution quickly. To solve the follower's problem efficiently, we apply some efficient deterministic optimization algorithms in the MATLAB Toolbox to search for its solutions. The asymptotically global convergence of the algorithm is proved. Numerical experiments on 25 test problems show that the proposed algorithm has a better performance than the compared algorithms on most of the test problems and is effective and efficient.

ISGP: Impedance Spectroscopy Analysis Using Evolutionary Programming Procedure

Abstract: Impedance spectroscopy measurement results in a complex function over several orders of magnitude of frequencies. The underlying physical phenomena are related to the relaxation times in the sample. The inverse problem of finding the distribution function of relaxation times is a demanding one. We have developed a modified Genetic Programming method for this task. The evolution force is composed of lowering the discrepancy between the model's prediction and the measured data, while keeping the model simple in terms of the number of free parameters. The program seeks distribution that has the form of a peak or a sum of several peaks, assuming the Debye kernel. All the peaks are known functions. By finding a functional form of the distribution, one may develop a physical model and examine its behavior. The different peaks, which can be related to different processes, can be analyzed separately.