Showing papers on "Evolutionary programming published in 2010"
TL;DR: Experimental results show that the performance of ECHT is better than each single constraint handling method used to form the ensemble with the respective EA, and competitive to the state-of-the-art algorithms.
Abstract: During the last three decades, several constraint handling techniques have been developed to be used with evolutionary algorithms (EAs). According to the no free lunch theorem, it is impossible for a single constraint handling technique to outperform all other techniques on every problem. In other words, depending on several factors such as the ratio between feasible search space and the whole search space, multimodality of the problem, the chosen EA, and global exploration/local exploitation stages of the search process, different constraint handling methods can be effective during different stages of the search process. Motivated by these observations, we propose an ensemble of constraint handling techniques (ECHT) to solve constrained real-parameter optimization problems, where each constraint handling method has its own population. A distinguishing feature of the ECHT is the usage of every function call by each population associated with each constraint handling technique. Being a general concept, the ECHT can be realized with any existing EA. In this paper, we present two instantiations of the ECHT using four constraint handling methods with the evolutionary programming and differential evolution as the EAs. Experimental results show that the performance of ECHT is better than each single constraint handling method used to form the ensemble with the respective EA, and competitive to the state-of-the-art algorithms.
399 citations
TL;DR: The generalized evolutionary framework focuses on attaining reliable search performance in the surrogate-assisted evolutionary framework by working on two major issues: to mitigate the 'curse of uncertainty' robustly, and to benefit from the 'bless of uncertainty.'
Abstract: Using surrogate models in evolutionary search provides an efficient means of handling today's complex applications plagued with increasing high-computational needs. Recent surrogate-assisted evolutionary frameworks have relied on the use of a variety of different modeling approaches to approximate the complex problem landscape. From these recent studies, one main research issue is with the choice of modeling scheme used, which has been found to affect the performance of evolutionary search significantly. Given that theoretical knowledge available for making a decision on an approximation model a priori is very much limited, this paper describes a generalization of surrogate-assisted evolutionary frameworks for optimization of problems with objectives and constraints that are computationally expensive to evaluate. The generalized evolutionary framework unifies diverse surrogate models synergistically in the evolutionary search. In particular, it focuses on attaining reliable search performance in the surrogate-assisted evolutionary framework by working on two major issues: 1) to mitigate the 'curse of uncertainty' robustly, and 2) to benefit from the 'bless of uncertainty.' The backbone of the generalized framework is a surrogate-assisted memetic algorithm that conducts simultaneous local searches using ensemble and smoothing surrogate models, with the aims of generating reliable fitness prediction and search improvements simultaneously. Empirical study on commonly used optimization benchmark problems indicates that the generalized framework is capable of attaining reliable, high quality, and efficient performance under a limited computational budget.
375 citations
TL;DR: In this article, a quantum-inspired particle swarm optimization (QPSO) is proposed, which has stronger search ability and quicker convergence speed, not only because of the introduction of quantum computing theory, but also due to two special implementations: self-adaptive probability selection and chaotic sequences mutation.
Abstract: Economic load dispatch (ELD) is an important topic in the operation of power plants which can help to build up effective generating management plans. The ELD problem has nonsmooth cost function with equality and inequality constraints which make it difficult to be effectively solved. Different heuristic optimization methods have been proposed to solve this problem in previous study. In this paper, quantum-inspired particle swarm optimization (QPSO) is proposed, which has stronger search ability and quicker convergence speed, not only because of the introduction of quantum computing theory, but also due to two special implementations: self-adaptive probability selection and chaotic sequences mutation. The proposed approach is tested with five standard benchmark functions and three power system cases consisting of 3, 13, and 40 thermal units. Comparisons with similar approaches including the evolutionary programming (EP), genetic algorithm (GA), immune algorithm (IA), and other versions of particle swarm optimization (PSO) are given. The promising results illustrate the efficiency of the proposed method and show that it could be used as a reliable tool for solving ELD problems.
288 citations
24 Nov 2010
TL;DR: This book introduces the new experimentalism in evolutionary computation, providing tools to understand algorithms and programs and their interaction with optimization problems, and provides a self-contained experimental methodology and many examples.
Abstract: This book introduces the new experimentalism in evolutionary computation, providing tools to understand algorithms and programs and their interaction with optimization problems. It develops and applies statistical techniques to analyze and compare modern search heuristics such as evolutionary algorithms and particle swarm optimization. The book bridges the gap between theory and experiment by providing a self-contained experimental methodology and many examples.
263 citations
Book•
05 Jan 2010
TL;DR: This comprehensive book explains how to use MATLAB to implement CI techniques for the solution of biological problems, and will help readers with their work on evolution dynamics, self-organization, natural and artificial morphogenesis, emergent collective behaviors, swarm intelligence, evolutionary strategies, genetic programming, and the evolution of social behaviors.
Abstract: Offering a wide range of programming examples implemented in MATLAB, Computational Intelligence Paradigms: Theory and Applications Using MATLAB presents theoretical concepts and a general framework for computational intelligence (CI) approaches, including artificial neural networks, fuzzy systems, evolutionary computation, genetic algorithms and programming, and swarm intelligence. It covers numerous intelligent computing methodologies and algorithms used in CI research. The book first focuses on neural networks, including common artificial neural networks; neural networks based on data classification, data association, and data conceptualization; and real-world applications of neural networks. It then discusses fuzzy sets, fuzzy rules, applications of fuzzy systems, and different types of fused neuro-fuzzy systems, before providing MATLAB illustrations of ANFIS, classification and regression trees, fuzzy c-means clustering algorithms, fuzzy ART map, and TakagiSugeno inference systems. The authors also describe the history, advantages, and disadvantages of evolutionary computation and include solved MATLAB programs to illustrate the implementation of evolutionary computation in various problems. After exploring the operators and parameters of genetic algorithms, they cover the steps and MATLAB routines of genetic programming. The final chapter introduces swarm intelligence and its applications, particle swarm optimization, and ant colony optimization. Full of worked examples and end-of-chapter questions, this comprehensive book explains how to use MATLAB to implement CI techniques for the solution of biological problems. It will help readers with their work on evolution dynamics, self-organization, natural and artificial morphogenesis, emergent collective behaviors, swarm intelligence, evolutionary strategies, genetic programming, and the evolution of social behaviors.
238 citations
Book•
01 Jan 2010
TL;DR: Multiobjective Optimization, Models and Applications.
Abstract: Multiobjective Optimization, Models and Applications- A Novel Smart Multi-Objective Particle Swarm Optimisation Using Decomposition- A Hybrid Scalarization and Adaptive ?-Ranking Strategy for Many-Objective Optimization- pMODE-LD+SS: An Effective and Efficient Parallel Differential Evolution Algorithm for Multi-Objective Optimization- Improved Dynamic Lexicographic Ordering for Multi-Objective Optimisation- Privacy-Preserving Multi-Objective Evolutionary Algorithms- Optimizing Delivery Time in Multi-Objective Vehicle Routing Problems with Time Windows- Speculative Evaluation in Particle Swarm Optimization- Towards Directed Open-Ended Search by a Novelty Guided Evolution Strategy- Consultant-Guided Search Algorithms with Local Search for the Traveling Salesman Problem- Many-Objective Test Problems to Visually Examine the Behavior of Multiobjective Evolution in a Decision Space- Preference-Based Multi-Objective Particle Swarm Optimization Using Desirabilities- GPGPU-Compatible Archive Based Stochastic Ranking Evolutionary Algorithm (G-ASREA) for Multi-Objective Optimization- Hybrid Directional-Biased Evolutionary Algorithm for Multi-Objective Optimization- A Framework for Incorporating Trade-Off Information Using Multi-Objective Evolutionary Algorithms- Applications, Engineering and Economical Models- Topography-Aware Sensor Deployment Optimization with CMA-ES- Evolutionary Optimization on Problems Subject to Changes of Variables- On-Line Purchasing Strategies for an Evolutionary Algorithm Performing Resource-Constrained Optimization- Parallel Artificial Immune System in Optimization and Identification of Composite Structures- Bioreactor Control by Genetic Programming- Solving the One-Commodity Pickup and Delivery Problem Using an Adaptive Hybrid VNS/SA Approach- Testing the Dinosaur Hypothesis under Empirical Datasets- Fractal Gene Regulatory Networks for Control of Nonlinear Systems- An Effective Hybrid Evolutionary Local Search for Orienteering and Team Orienteering Problems with Time Windows- Discrete Differential Evolution Algorithm for Solving the Terminal Assignment Problem- Decentralized Evolutionary Agents Streamlining Logistic Network Design- Testing the Permutation Space Based Geometric Differential Evolution on the Job-Shop Scheduling Problem- New Uncertainty Handling Strategies in Multi-objective Evolutionary Optimization- Evolving a Single Scalable Controller for an Octopus Arm with a Variable Number of Segments- Multi-agent Systems and Parallel Approaches- An Island Model for the No-Wait Flow Shop Scheduling Problem- Environment-Driven Embodied Evolution in a Population of Autonomous Agents- Large-Scale Global Optimization Using Cooperative Coevolution with Variable Interaction Learning- EvoShelf: A System for Managing and Exploring Evolutionary Data- Differential Evolution Algorithms with Cellular Populations- Flocking in Stationary and Non-stationary Environments: A Novel Communication Strategy for Heading Alignment- Evolution of XPath Lists for Document Data Selection- PMF: A Multicore-Enabled Framework for the Construction of Metaheuristics for Single and Multiobjective Optimization- Parallel Evolutionary Approach of Compaction Problem Using MapReduce- Ant Colony Optimization with Immigrants Schemes in Dynamic Environments- Secret Key Specification for a Variable-Length Cryptographic Cellular Automata Model- Variable Neighborhood Search and Ant Colony Optimization for the Rooted Delay-Constrained Minimum Spanning Tree Problem- Adaptive Modularization of the MAPK Signaling Pathway Using the Multiagent Paradigm- Genetic Computing and Games- Experimental Comparison of Methods to Handle Boundary Constraints in Differential Evolution- Entropy-Driven Evolutionary Approaches to the Mastermind Problem- Evolutionary Detection of New Classes of Equilibria: Application in Behavioral Games- Design and Comparison of two Evolutionary Approaches for Solving the Rubik's Cube- Statistical Analysis of Parameter Setting in Real-Coded Evolutionary Algorithms- Performance of Network Crossover on NK Landscapes and Spin Glasses- Promoting Phenotypic Diversity in Genetic Programming- A Genetic Programming Approach to the Matrix Bandwidth-Minimization Problem- Using Co-solvability to Model and Exploit Synergetic Effects in Evolution- Fast Grammar-Based Evolution Using Memoization- Evolution of Conventions and Social Polarization in Dynamical Complex Networks- Evolving Strategies for Updating Pheromone Trails: A Case Study with the TSP- The Role of Syntactic and Semantic Locality of Crossover in Genetic Programming- The Layered Learning Method and Its Application to Generation of Evaluation Functions for the Game of Checkers
175 citations
TL;DR: The fixation probability and the average fixation time are investigated not only up to linear but also up to higher orders in selection intensity, which finds universal higher order expansions, which allow a rescaling of the selection intensity.
Abstract: Weak selection, which means a phenotype is slightly advantageous over another, is an important limiting case in evolutionary biology. Recently, it has been introduced into evolutionary game theory. In evolutionary game dynamics, the probability to be imitated or to reproduce depends on the performance in a game. The influence of the game on the stochastic dynamics in finite populations is governed by the intensity of selection. In many models of both unstructured and structured populations, a key assumption allowing analytical calculations is weak selection, which means that all individuals perform approximately equally well. In the weak selection limit many different microscopic evolutionary models have the same or similar properties. How universal is weak selection for those microscopic evolutionary processes? We answer this question by investigating the fixation probability and the average fixation time not only up to linear but also up to higher orders in selection intensity. We find universal higher order expansions, which allow a rescaling of the selection intensity. With this, we can identify specific models which violate (linear) weak selection results, such as the one-third rule of coordination games in finite but large populations.
171 citations
TL;DR: With the proposed method, the performance metric has improved significantly and the speed of the parent selection process has also increased when compared with the non-domination sorting, and the proposed algorithm also outperforms ten other algorithms.
119 citations
23 Feb 2010
TL;DR: This book discusses the mutual intersection of two interesting fields of research, i.e. deterministic chaos and evolutionary computation, which are able to handle tasks such as control of various chaotic systems and synthesis of their structure.
Abstract: This book discusses the mutual intersection of two interesting fields of research, i.e. deterministic chaos and evolutionary computation. Evolutionary computation which are able to handle tasks such as control of various chaotic systems and synthesis of their structure are explored, while deterministic chaos is investigated as a behavioral part of evolutionary algorithms. This book is targeted for a number of audiences. Firstly, it will be an instructional material for senior undergraduate and entry-point graduate students in computer science, physics, applied mathematics, and engineering, who are working in the area of deterministic chaos and evolutionary algorithms. Secondly, researchers who desire to know how to apply evolutionary techniques on chaos control as well as researchers interested in the emergence of chaos in evolutionary algorithms will find this book a very useful tool and starting point. And finally, this book can be viewed as a resource handbook and material for practitioners who want to apply these methods that solve practical problems to their challenging applications.
119 citations
TL;DR: The performance of ensemble is compared with a mixed mutation strategy, which integrates several mutation operators into a single algorithm as well as against the recently proposed Adaptive EP using Gaussian and Cauchy mutations.
118 citations
TL;DR: In this article, a Gray CoRrelation Analysis (GCRA) method is proposed to integrate the objectives and provide a relative measure to a particular switching plan associated with a chromosome without any prior knowledge of the system under reconfiguration.
Abstract: Feeder reconfiguration is a common technique that is used by distribution system operators during normal or emergency operational planning. By changing the status of switches on the distribution systems, the feeders can be reconfigured. During a feeder reconfiguration, more than one objective is considered by the distribution system operators. Due to the complexity of the reconfiguration problems, the system operators are looking for assistance from computer program that can provide adequate switching plans to reconfigure the feeders such that the desired goal can be achieved. Thus, the feeder reconfiguration is a type of discrete multi-objective optimization problems. Evolutionary programming (EP) technique is a method that can be applied to identify an optimal switching plan for feeder reconfiguration. A fitness function is required in EP for chromosome selection during reproduction process. The fitness function needs to integrate the objectives to provide a measure for each chromosome. Normalizing the objectives is a typical method for multi-objective optimizations such that these objectives are comparable. In this paper, Gray CoRrelation Analysis (GCRA) method is proposed. The proposed method is used to integrate the objectives and provide a relative measure to a particular switching plan associated with a chromosome without any prior knowledge of the system under reconfiguration. Two different distribution systems are used in this paper to demonstrate how the proposed GCRA is applied during the selection process of EP. Several simulations show that the EP can identify the solution more accurately when GCRA is applied than other methods.
TL;DR: A survey of self-adaptive parameter control in evolutionary computation classifies self- Adaptation in the taxonomy of parameter setting techniques, gives an overview of automatic online-controllable evolutionary operators and provides a coherent view on search techniques in the space of strategy parameters.
Abstract: The success of evolutionary search depends on adequate parameter settings. Ill conditioned strategy parameters decrease the success probabilities of genetic operators. Proper settings may change during the optimization process. The question arises if adequate settings can be found automatically during the optimization process. Evolution strategies gave an answer to the online parameter control problem decades ago: self-adaptation. Self-adaptation is the implicit search in the space of strategy parameters. The self-adaptive control of mutation strengths in evolution strategies turned out to be exceptionally successful. Nevertheless, for years self-adaptation has not achieved the attention it deserves. This paper is a survey of self-adaptive parameter control in evolutionary computation. It classifies self-adaptation in the taxonomy of parameter setting techniques, gives an overview of automatic online-controllable evolutionary operators and provides a coherent view on search techniques in the space of strategy parameters. Beyer and Sendhoff’s covariance matrix self-adaptation evolution strategy is reviewed as a successful example for self-adaptation and exemplarily tested for various concepts that are discussed.
01 Mar 2010
TL;DR: The results of a study conducted to investigate the use of genetic algorithms (GAs) as a means of inducing solutions to the examination timetabling problem (ETP) show the performance of the system is comparable to that of other evolutionary techniques and in some cases the system was found to outperform these techniques.
Abstract: This paper presents the results of a study conducted to investigate the use of genetic algorithms (GAs) as a means of inducing solutions to the examination timetabling problem (ETP). This study differs from previous efforts applying genetic algorithms to this domain in that firstly it takes a two-phased approach to the problem which focuses on producing timetables that meet the hard constraints during the first phase, while improvements are made to these timetables in the second phase so as to reduce the soft constraint costs. Secondly, domain specific knowledge in the form of heuristics is used to guide the evolutionary process. The system was tested on a set of 13 real-world problems, namely, the Carter benchmarks. The performance of the system on the benchmarks is comparable to that of other evolutionary techniques and in some cases the system was found to outperform these techniques. Furthermore, the quality of the examination timetables evolved is within range of the best results produced in the field.
TL;DR: BBO algorithm is matured to optimize the element length and spacing for Yagi-Uda antenna and the results obtained are compared with the genetic algorithm (GA), evolutionary programming (EP), comprehensive learning particle swarm optimization (CLPSO), simulated annealing (SA) and computational intelligence (CI).
Abstract: Biogeography based optimization (BBO) is a new inclusive vigor based on the science of biogeography. Biogeography is the schoolwork of geographical allotment of biological organisms. BBO employs migration operator to share information between the problem solutions. The problem solutions are identified as habitat and sharing of features is called migration. In this communication, BBO algorithm is matured to optimize the element length and spacing for Yagi-Uda antenna. The gain of Yagi-Uda is a multimodal function and is hard to optimize because of its reliance on changes in lengths and spacings. To confirm the capabilities of BBO, Yagi-Uda antenna is optimized for three different design objectives which include gain, input impedance and side lobe level (SLL). During optimization, NEC2-a method of moment's code evaluates the performance of each design generated by BBO algorithm. The results obtained by BBO are compared with the genetic algorithm (GA), evolutionary programming (EP), comprehensive learning particle swarm optimization (CLPSO), simulated annealing (SA) and computational intelligence (CI).
TL;DR: In this paper, the authors presented differential evolution to solve the combined heat and power economic dispatch problem, which is an improved version of the genetic algorithm and evolutionary programming, is a very simple, fast, and robust global optimization technique.
Abstract: This article presents differential evolution to solve the combined heat and power economic dispatch problem. Differential evolution, an improved version of the genetic algorithm and evolutionary programming, is a very simple, fast, and robust global optimization technique. The proposed algorithm is illustrated for a test system, and the test results are compared with those obtained from particle swarm optimization and evolutionary programming.
TL;DR: An algorithm is presented that overcomes a perceived limitation of evolutionary art and design algorithms by employing an automatic fitness function and notes that in human creativity, change is less choosing amongst randomly generated variants and more capitalizing on the associative structure of a conceptual network to hone in on a vision.
Abstract: A perceived limitation of evolutionary art and design algorithms is that they rely on human intervention; the artist selects the most aesthetically pleasing variants of one generation to produce the next. This paper discusses how computer generated art and design can become more creatively human-like with respect to both process and outcome. As an example of a step in this direction, we present an algorithm that overcomes the above limitation by employing an automatic fitness function. The goal is to evolve abstract portraits of Darwin, using our 2nd generation fitness function which rewards genomes that not just produce a likeness of Darwin but exhibit certain strategies characteristic of human artists. We note that in human creativity, change is less choosing amongst randomly generated variants and more capitalizing on the associative structure of a conceptual network to hone in on a vision. We discuss how to achieve this fluidity algorithmically.
Journal Article•
TL;DR: Among the various CCAA designs, the design containing central element and 4, 6 and 8 elements in three successive concentric rings proves to be such global optimal design set with global minimum SLL (−39.66 dB) as determined by Evolutionary Programming.
Abstract: In this paper the maximum sidelobe level (SLL) reductions without and with central element feeding in various designs of three-ring concentric circular antenna arrays (CCAA) are examined using a real-coded Evolutionary Programming (EP) to finally determine the global optimal three-ring CCAA design. Standard real-coded Particle Swarm Optimization (PSO) and real-coded Particle Swarm Optimization with Constriction Factor and Inertia Weight Approach (PSOCFIWA) are also employed for comparative optimization but both prove to be suboptimal. This paper assumes non-uniform excitation weights and uniform spacing of excitation elements in each three-ring CCAA design. Among the various CCAA designs, the design containing central element and 4, 6 and 8 elements in three successive concentric rings proves to be such global optimal design set with global minimum SLL (−39.66 dB) as determined by Evolutionary Programming.
TL;DR: This paper investigates evolutionary games with the invasion process updating rules on three simple non-directed graphs: the star, the circle and the complete graph and derives the exact solutions of the stochastic evolutionary game dynamics.
Abstract: In this paper, we investigate evolutionary games with the invasion process updating rules on three simple non-directed graphs: the star, the circle and the complete graph. Here, we present an analytical approach and derive the exact solutions of the stochastic evolutionary game dynamics. We present formulae for the fixation probability and also for the speed of the evolutionary process, namely for the mean time to absorption (either mutant fixation or extinction) and then the mean time to mutant fixation. Through numerical examples, we compare the different impact of the population size and the fitness of each type of individual on the above quantities on the three different structures. We do this comparison in two specific cases. Firstly, we consider the case where mutants have fixed fitness r and resident individuals have fitness 1. Then, we consider the case where the fitness is not constant but depends on games played among the individuals, and we introduce a ‘hawk–dove’ game as an example.
TL;DR: In this paper, the maximum sidelobe level (SLL) reductions without and with central element feeding in various designs of three-ring concentric circular antenna arrays (CCAA) are examined using a real-coded Evolutionary Programming (EP) to finally determine the global optimal three ring CCAA design.
Abstract: In this paper the maximum sidelobe level (SLL) reductions without and with central element feeding in various designs of three-ring concentric circular antenna arrays (CCAA) are examined using a real-coded Evolutionary Programming (EP) to finally determine the global optimal three-ring CCAA design. Standard real-coded Particle Swarm Optimization (PSO) and real-coded Particle Swarm Optimization with Constriction Factor and Inertia Weight Approach (PSOCFIWA) are also employed for comparative optimization but both prove to be suboptimal. This paper assumes non-uniform excitation weights and uniform spacing of excitation elements in each three-ring CCAA design. Among the various CCAA designs, the design containing central element and 4, 6 and 8 elements in three successive concentric rings proves to be such global optimal design set with global minimum SLL (−39.66 dB) as determined by Evolutionary Programming.
01 Jan 2010
TL;DR: The techniques used in evolution-based gait optimization are reviewed, including why Evolutionary Computation techniques should be used, how fitness functions should be composed, and the selection of genetic operators and control parameters.
Abstract: Gait generation is very important as it directly affects the quality of locomotion of legged robots. As this is an optimization problem with constraints, it readily lends itself to Evolutionary Computation methods and solutions. This paper reviews the techniques used in evolution-based gait optimization, including why Evolutionary Computation techniques should be used, how fitness functions should be composed, and the selection of genetic operators and control parameters. This paper also addresses further possible improvements in the efficiency and quality of evolutionary gait optimization, some problems that have not yet been resolved and the perspectives for related future research.
TL;DR: This paper discusses the application of evolutionary programming methods to the problem of analyzing impedance spectroscopy results, and two complementary methods have been applied: Genetic Algorithm and Genetic Programming (GP).
Abstract: This paper discusses the application of evolutionary programming methods to the problem of analyzing impedance spectroscopy results. The basic approach is a “direct-problem” one, i.e., to find a time constant distribution function that would create similar impedance results as the measured ones, within experimental error. Two complementary methods have been applied and are discussed here: Genetic Algorithm (GA) and Genetic Programming (GP). A GA can be applied when a known (or desired) model exists, whereas GP can be used to create new models where the only a-priori knowledge is their smoothness and their non-negativity. GP is tuned to prefer relatively non-complex models through penalization of unnecessary complexity.
TL;DR: In this paper, a comparative study has been made on the solutions obtained using combined economic emission dispatch (CEED) problem considering line flow constraints using different intelligent techniques for the regulated power system to ensure a practical, economical and secure generation schedule.
Book•
25 Nov 2010
TL;DR: This paper presents Parallel Ant Colony Optimization for 3D Protein Structure Prediction using the HP Lattice Model, a model for Parallel Operators in Genetic Algorithms.
Abstract: Parallel Evolutionary Optimization.- A Model for Parallel Operators in Genetic Algorithms.- Parallel Evolutionary Multiobjective Optimization.- Parallel Hardware for Genetic Algorithms.- A Reconfigurable Parallel Hardware for Genetic Algorithms.- Reconfigurable Computing and Parallelism for Implementing and Accelerating Evolutionary Algorithms.- Distributed Evolutionary Computation.- Performance of Distributed GAs on DNA Fragment Assembly.- On Parallel Evolutionary Algorithms on the Computational Grid.- Parallel Evolutionary Algorithms on Consumer-Level Graphics Processing Unit.- Parallel Particle Swarm Optimization.- Intelligent Parallel Particle Swarm Optimization Algorithms.- Parallel Ant Colony Optimization for 3D Protein Structure Prediction using the HP Lattice Model.
07 Jul 2010
TL;DR: Different variations of the classical (1+1) evolutionary algorithm, all imitating the property that the (1-1) EA over intervals of time touches all bits roughly the same number of times are proposed.
Abstract: Motivated by recent successful applications of the concept of quasirandomness, we investigate to what extent such ideas can be used in evolutionary computation. To this aim, we propose different variations of the classical (1+1) evolutionary algorithm, all imitating the property that the (1+1) EA over intervals of time touches all bits roughly the same number of times. We prove bounds on the optimization time of these algorithms for the simple OneMax function.Surprisingly, none of the algorithms achieves the seemingly obvious reduction of the runtime from Θ(n log n) to O(n). On the contrary, one may even need Ω(n2) time. However, we also find that quasirandom ideas, if implemented correctly, can yield an over 50% speed-up.
TL;DR: It is hoped that the findings put forward here can be used to design computational models of evolution that produce significant gains in evolvability and evolutionary speed.
Abstract: Biological and artificial evolutionary systems exhibit varying degrees of evolvability and different rates of evolution. Such quantities can be affected by various factors. Here, we review some evolutionary mechanisms and discuss new developments in biology that can potentially improve evolvability or accelerate evolution in artificial systems. Biological notions are discussed to the degree they correspond to notions in Evolutionary Computation. We hope that the findings put forward here can be used to design computational models of evolution that produce significant gains in evolvability and evolutionary speed.
TL;DR: A new evolutionary design tool is presented, which uses shape grammars and a grammar-based form of evolutionary computation, grammatical evolution (GE), which allows forms to be iteratively selected.
Abstract: A new evolutionary design tool is presented, which uses shape grammars and a grammar-based form of evolutionary computa- tion, grammatical evolution (GE). Shape grammars allow the user to specify possible forms, and GE allows forms to be iteratively selected,
TL;DR: An Evolutionary Memetic Algorithm (EMA), which uses a local search intensity scheme to complement the global search capability of Evolutionary Algorithms (EAs), is proposed for rule extraction.
Abstract: In this paper, an Evolutionary Memetic Algorithm (EMA), which uses a local search intensity scheme to complement the global search capability of Evolutionary Algorithms (EAs), is proposed for rule extraction. Two schemes for local search are studied, namely [email protected], which uses a micro-Genetic Algorithm-based (@mGA) technique, and EMA-AIS, which is inspired by Artificial Immune System (AIS) and uses the clonal selection for cell proliferation. The evolutionary memetic algorithm is complemented with the use of a variable-length chromosome structure, which allows the flexibility to model the number of rules required. In addition, advanced variation operators are used to improve different aspects of the algorithm. Real world benchmarking problems are used to validate the performance of EMA and results from simulations show the proposed algorithm is effective.
01 Jan 2010
TL;DR: A hybrid algorithm of DE and EP, denoted as DEEP, is proposed in this study to reduce the required population size and is designed as a novel primary-auxiliary model to minimise the additional computational cost.
Abstract: Differential evolution (DE) is a promising evolutionary algorithm for solving the optimal reactive power flow (ORPF) problem, but it requires relatively large population size to avoid premature convergence, which will increase the computational time. On the other hand, evolutionary programming (EP) has been proved to have good global search ability. Exploiting this complementary feature, a hybrid algorithm of DE and EP, denoted as DEEP, is proposed in this study to reduce the required population size. The hybridisation is designed as a novel primary-auxiliary model to minimise the additional computational cost. The effectiveness of DEEP is verified by the serial simulations on the IEEE 14-, 30-, 57-bus system test cases and the parallel simulations on the IEEE 118-bus system test case.
TL;DR: An evolutionary algorithm that improves the efficiency of the standard genetic algorithm by considering cooperation with a local search around some of the solutions it visits and an approach based on simulated annealing that uses the same representation scheme of a feasible solution to solve the cell formation problem.
Abstract: The cell formation problem is a crucial component of a cell production design in a manufacturing system. This problem consists of a set of product parts to be manufactured in a group of machines. The objective is to build manufacturing clusters by associating part families with machine cells, with the aim of minimizing the inter-cellular movements of parts by grouping efficacy measures. We present two approaches to solve the cell formation problem. First, we present an evolutionary algorithm that improves the efficiency of the standard genetic algorithm by considering cooperation with a local search around some of the solutions it visits. Second, we present an approach based on simulated annealing that uses the same representation scheme of a feasible solution. To evaluate the performance of both algorithms, we used a known set of CFP instances. We compared the results of both algorithms with the results of five other algorithms from the literature. In eight out of 36 instances we considered, the evolutionary method outperformed the previous results of other evolutionary algorithms, and in 26 instances it found the same best solutions. On the other hand, simulated annealing not only found the best previously known solutions, but it also found better solutions than existing ones for various problems.