Showing papers on "Crossover published in 1999"
••
01 Jun 1999TL;DR: A novel hybrid genetic algorithm that finds a globally optimal partition of a given data into a specified number of clusters using a classical gradient descent algorithm used in clustering, viz.
Abstract: In this paper, we propose a novel hybrid genetic algorithm (GA) that finds a globally optimal partition of a given data into a specified number of clusters. GA's used earlier in clustering employ either an expensive crossover operator to generate valid child chromosomes from parent chromosomes or a costly fitness function or both. To circumvent these expensive operations, we hybridize GA with a classical gradient descent algorithm used in clustering, viz. K-means algorithm. Hence, the name genetic K-means algorithm (GKA). We define K-means operator, one-step of K-means algorithm, and use it in GKA as a search operator instead of crossover. We also define a biased mutation operator specific to clustering called distance-based-mutation. Using finite Markov chain theory, we prove that the GKA converges to the global optimum. It is observed in the simulations that GKA converges to the best known optimum corresponding to the given data in concurrence with the convergence result. It is also observed that GKA searches faster than some of the other evolutionary algorithms used for clustering.
1,326 citations
••
TL;DR: The compact genetic algorithm (cGA) is introduced which represents the population as a probability distribution over the set of solutions and is operationally equivalent to the order-one behavior of the simple GA with uniform crossover.
Abstract: Introduces the compact genetic algorithm (cGA) which represents the population as a probability distribution over the set of solutions and is operationally equivalent to the order-one behavior of the simple GA with uniform crossover. It processes each gene independently and requires less memory than the simple GA. The development of the compact GA is guided by a proper understanding of the role of the GA's parameters and operators. The paper clearly illustrates the mapping of the simple GA's parameters into those of an equivalent compact GA. Computer simulations compare both algorithms in terms of solution quality and speed. Finally, this work raises important questions about the use of information in a genetic algorithm, and its ramifications show us a direction that can lead to the design of more efficient GAs.
1,049 citations
••
TL;DR: This paper presents crossover and mutation operators, developed to tackle the Travelling Salesman Problem with Genetic Algorithms with different representations such as: binary representation, path representation, adjacency representation, ordinal representation and matrix representation.
Abstract: This paper is the result of a literature study carried out by the authors. It is a review of the different attempts made to solve the Travelling Salesman Problem with Genetic Algorithms. We present crossover and mutation operators, developed to tackle the Travelling Salesman Problem with Genetic Algorithms with different representations such as: binary representation, path representation, adjacency representation, ordinal representation and matrix representation. Likewise, we show the experimental results obtained with different standard examples using combination of crossover and mutation operators in relation with path representation.
839 citations
•
25 Feb 1999
TL;DR: Genetic Algorithms in Speech Recognition Systems.- 8.2.1 Background of Speech Recogning Systems.
Abstract: 1. Introduction, Background and Biological Inspiration.- 1.1 Biological Background.- 1.1.1 Coding of DNA.- 1.1.2 Flow of Genetic Information.- 1.1.3 Recombination.- 1.1.4 Mutation.- 1.2 Conventional Genetic Algorithm.- 1.3 Theory and Hypothesis.- 1.3.1 Schema Theory.- 1.3.2 Building Block Hypothesis.- 1.4 A Simple Example.- 2. Modifications to Genetic Algorithms.- 2.1 Chromosome Representation.- 2.2 Objective and Fitness Functions.- 2.2.1 Linear Scaling.- 2.2.2 Sigma Truncation.- 2.2.3 Power Law Scaling.- 2.2.4 Ranking.- 2.3 Selection Methods.- 2.4 Genetic Operations.- 2.4.1 Crossover.- 2.4.2 Mutation.- 2.4.3 Operational Rates Settings.- 2.4.4 Reordering.- 2.5 Replacement Scheme.- 2.6 A Game of Genetic Creatures.- 2.7 Chromosome Representation.- 2.8 Fitness Function.- 2.9 Genetic Operation.- 2.9.1 Selection Window for Functions and Parameters.- 2.10 Demo and Run.- 3. Intrinsic Characteristics.- 3.1 Parallel Genetic Algorithm.- 3.1.1 Global GA.- 3.1.2 Migration GA.- 3.1.3 Diffusion GA.- 3.2 Multiple Objective.- 3.3 Robustness.- 3.4 Multimodal.- 3.5 Constraints.- 3.5.1 Searching Domain.- 3.5.2 Repair Mechanism.- 3.5.3 Penalty Scheme.- 3.5.4 Specialized Genetic Operations.- 4. Hierarchical Genetic Algorithm.- 4.1 Biological Inspiration.- 4.1.1 Regulatory Sequences and Structural Genes.- 4.1.2 Active and Inactive Genes.- 4.2 Hierarchical Chromosome Formulation.- 4.3 Genetic Operations.- 4.4 Multiple Objective Approach.- 4.4.1 Iterative Approach.- 4.4.2 Group Technique.- 4.4.3 Multiple-Objective Ranking.- 5. Genetic Algorithms in Filtering.- 5.1 Digital IIR Filter Design.- 5.1.1 Chromosome Coding.- 5.1.2 The Lowest Filter Order Criterion.- 5.2 Time Delay Estimation.- 5.2.1 Problem Formulation.- 5.2.2 Genetic Approach.- 5.2.3 Results.- 5.3 Active Noise Control.- 5.3.1 Problem Formulation.- 5.3.2 Simple Genetic Algorithm.- 5.3.3 Multiobjective Genetic Algorithm Approach.- 5.3.4 Parallel Genetic Algorithm Approach.- 5.3.5 Hardware GA Processor.- 6. Genetic Algorithms in H-infinity Control.- 6.1 A Mixed Optimization Design Approach.- 6.1.1 Hierarchical Genetic Algorithm.- 6.1.2 Application I: The Distillation Column Design.- 6.1.3 Application II: Benchmark Problem.- 6.1.4 Design Comments.- 7. Hierarchical Genetic Algorithms in Computational Intelligence.- 7.1 Neural Networks.- 7.1.1 Introduction of Neural Network.- 7.1.2 HGA Trained Neural Network (HGANN).- 7.1.3 Simulation Results.- 7.1.4 Application of HGANN on Classification.- 7.2 Fuzzy Logic.- 7.2.1 Basic Formulation of Fuzzy Logic Controller.- 7.2.2 Hierarchical Structure.- 7.2.3 Application I: Water Pump System.- 7.2.4 Application II: Solar Plant.- 8. Genetic Algorithms in Speech Recognition Systems.- 8.1 Background of Speech Recognition Systems.- 8.2 Block Diagram of a Speech Recognition System.- 8.3 Dynamic Time Warping.- 8.4 Genetic Time Warping Algorithm (GTW).- 8.4.1 Encoding mechanism.- 8.4.2 Fitness function.- 8.4.3 Selection.- 8.4.4 Crossover.- 8.4.5 Mutation.- 8.4.6 Genetic Time Warping with Relaxed Slope Weighting Function (GTW-RSW).- 8.4.7 Hybrid Genetic Algorithm.- 8.4.8 Performance Evaluation.- 8.5 Hidden Markov Model using Genetic Algorithms.- 8.5.1 Hidden Markov Model.- 8.5.2 Training Discrete HMMs using Genetic Algorithms.- 8.5.3 Genetic Algorithm for Continuous HMM Training.- 8.6 A Multiprocessor System for Parallel Genetic Algorithms.- 8.6.1 Implementation.- 8.7 Global GA for Parallel GA-DTW and PGA-HMM.- 8.7.1 Experimental Results of Nonlinear Time-Normalization by the Parallel GA-DTW.- 8.8 Summary.- 9. Genetic Algorithms in Production Planning and Scheduling Problems.- 9.1 Background of Manufacturing Systems.- 9.2 ETPSP Scheme.- 9.2.1 ETPSP Model.- 9.2.2 Bottleneck Analysis.- 9.2.3 Selection of Key-Processes.- 9.3 Chromosome Configuration.- 9.3.1 Operational Parameters for GA Cycles.- 9.4 GA Application for ETPSP.- 9.4.1 Case 1: Two-product ETPSP.- 9.4.2 Case 2: Multi-product ETPSP.- 9.4.3 Case 3: MOGA Approach.- 9.5 Concluding Remarks.- 10. Genetic Algorithms in Communication Systems.- 10.1 Virtual Path Design in ATM.- 10.1.1 Problem Formulation.- 10.1.2 Average packet delay.- 10.1.3 Constraints.- 10.1.4 Combination Approach.- 10.1.5 Implementation.- 10.1.6 Results.- 10.2 Mesh Communication Network Design.- 10.2.1 Design of Mesh Communication Networks.- 10.2.2 Network Optimization using GA.- 10.2.3 Implementation.- 10.2.4 Results.- 10.3 Wireles Local Area Network Design.- 10.3.1 Problem Formulation.- 10.3.2 Multiobjective HGA Approach.- 10.3.3 Implementation.- 10.3.4 Results.- Appendix A.- Appendix B.- Appendix C.- Appendix D.- Appendix E.- Appendix F.- References.
626 citations
••
TL;DR: The results demonstrate that a genetic algorithm could be satisfactorily used in real time operations with stochastically generated inflows and the known global optimum for the four-reservoir problem can be achieved with real-value coding.
Abstract: Several alternative formulations of a genetic algorithm for reservoir systems are evaluated using the four-reservoir, deterministic, finite-horizon problem. This has been done with a view to presenting fundamental guidelines for implementation of the approach to practical problems. Alternative representation, selection, crossover, and mutation schemes are considered. It is concluded that the most promising genetic algorithm approach for the four-reservoir problem comprises real-value coding, tournament selection, uniform crossover, and modified uniform mutation. The real-value coding operates significantly faster than binary coding and produces better results. The known global optimum for the four-reservoir problem can be achieved with real-value coding. A nonlinear four-reservoir problem is considered also, along with one with extended time horizons. The results demonstrate that a genetic algorithm could be satisfactorily used in real time operations with stochastically generated inflows. A more complex ...
478 citations
•
13 Jul 1999TL;DR: Experimental results using test functions showed SPX works well on functions having multimodality and/or epistasis with a medium number of parents: 3-parent on a low dimensional function or 4 parents on high dimensional functions.
Abstract: In this paper, we proposed simplex crossover (SPX), a multi-parent recombination operator for real-coded genetic algorithms. SPX generates offspring vector values by uniformly sampling values from simplex formed by m (2 ≤ m ≤ number of parameters + 1) parent vectors. The SPX features an independence from of coordinate systems. Experimental results using test functions, which are commonly used in studies of evolutionary algorithms, showed SPX works well on functions having multimodality and/or epistasis with a medium number of parents: 3-parent on a low dimensional function or 4 parents on high dimensional functions.
363 citations
•
13 Jul 1999TL;DR: This paper explores the development of a GA that fulfills this requirement, and takes into account several aspects of the theory of GAs, including previous research work on population sizing, the schema theorem, building block mixing, and genetic drift.
Abstract: From the user's point of view, setting the parameters of a genetic algorithm (GA) is far from a trivial task. Moreover, the user is typically not interested in population sizes, crossover probabilities, selection rates, and other GA technicalities. He is just interested in solving a problem, and what he would really like to do, is to hand-in the problem to a blackbox algorithm, and simply press a start button. This paper explores the development of a GA that fulfills this requirement. It has no parameters whatsoever. The development of the algorithm takes into account several aspects of the theory of GAs, including previous research work on population sizing, the schema theorem, building block mixing, and genetic drift.
294 citations
••
01 Apr 1999TL;DR: This paper presents a neuro-fuzzy logic controller where all of its parameters can be tuned simultaneously by GA, and shows that the proposed controller offers encouraging advantages and has better performance.
Abstract: Due to their powerful optimization property, genetic algorithms (GAs) are currently being investigated for the development of adaptive or self-tuning fuzzy logic control systems. This paper presents a neuro-fuzzy logic controller (NFLC) where all of its parameters can be tuned simultaneously by GA. The structure of the controller is based on the radial basis function neural network (RBF) with Gaussian membership functions. The NFLC tuned by GA can somewhat eliminate laborious design steps such as manual tuning of the membership functions and selection of the fuzzy rules. The GA implementation incorporates dynamic crossover and mutation probabilistic rates for faster convergence. A flexible position coding strategy of the NFLC parameters is also implemented to obtain near optimal solutions. The performance of the proposed controller is compared with a conventional fuzzy controller and a PID controller tuned by GA. Simulation results show that the proposed controller offers encouraging advantages and has better performance.
256 citations
••
TL;DR: A flexible approach using the genetic algorithm (GA) is proposed for array failure correction in digital beamforming of arbitrary arrays, and three mating schemes, adjacent-fitness-paring, best-mate-worst, and emperor-selective are proposed and their performances are studied.
Abstract: A flexible approach using the genetic algorithm (GA) is proposed for array failure correction in digital beamforming of arbitrary arrays. In this approach, beamforming weights of an array are represented directly by a vector of complex numbers. The decimal linear crossover is employed so that no binary coding and decoding is necessary. Three mating schemes, adjacent-fitness-paring (AFP), best-mate-worst (BMW), and emperor-selective (EMS), are proposed and their performances are studied. Near-solutions from other analytic or heuristic techniques may be injected into the initial population to speed up convergence. Numerical examples of single- and multiple-element failure correction are presented to show the effectiveness of the approach.
234 citations
••
TL;DR: A new approach is presented that uses a small population of SA runs in a genetic algorithm (GA) framework and yields excellent results on the classical test examples of the JSP.
210 citations
••
TL;DR: A new combined approach, where a genetic algorithm is improved with the introduction of some knowledge about the scheduling problem represented by the use of a list heuristic in the crossover and mutation genetic operations, which shows that the knowledge-augmented algorithm produces much better results in terms of quality of solutions, although being slower interms of execution time.
Abstract: In the multiprocessor scheduling problem, a given program is to be scheduled in a given multiprocessor system such that the program's execution time is minimized. This problem being very hard to solve exactly, many heuristic methods for finding a suboptimal schedule exist. We propose a new combined approach, where a genetic algorithm is improved with the introduction of some knowledge about the scheduling problem represented by the use of a list heuristic in the crossover and mutation genetic operations. This knowledge-augmented genetic approach is empirically compared with a "pure" genetic algorithm and with a "pure" list heuristic, both from the literature. Results of the experiments carried out with synthetic instances of the scheduling problem show that our knowledge-augmented algorithm produces much better results in terms of quality of solutions, although being slower in terms of execution time.
••
TL;DR: Results of tests indicate that genetic algorithms provide an efficient algorithm for PMSP_E/T; that neighborhood exchange type of search can yield relatively better results in small and easy instances of the problem but the genetic algorithm with the crossover operator outperforms such search in larger-sized, more difficult problems; and that the recombinative power of the genetic algorithms with therossover operator improves with increasing problem size and difficulty making it ever more attractive for applications of larger sizes.
01 Nov 1999
••
06 Jul 1999TL;DR: This work explores the effects of allowing mutation operators, crossover operators, and their associated rates to vary under the influence of selection on the suitability of alternative mutation models in dynamic landscapes.
Abstract: Evolvability refers to the adaptation of a population's genetic operator set over time. In traditional genetic algorithms, the genetic operator set, consisting of mutation operators, crossover operators, and their associated rates, is usually fixed. We explore the effects of allowing these operators and rates to vary under the influence of selection. The paper focuses on the suitability of alternative mutation models in dynamic landscapes. The mutation models include both traditional models in which all members of the population are subject to the same level of mutation and models in which mutation rates are genetically controlled.
••
01 Jan 1999TL;DR: The chapter considers the ‘permutations’ problem and introduces the concept of ‘shift’, which is used to train a neural network as an alternative to back-propagation, and considers the implicit parallelism of the GA.
Abstract: The basic genetic algorithm is introduced including the representation of individuals in populations, data structures for the representation of variables, binary strings, assessment of individual fitness, selection for recombination, crossover and mutation operators. The penalty function method of handling design constraints is introduced. The basic GA is illustrated by optimizing a simple structural design. We consider how we might improve the GA by on-line adaptation of the main controls. We then review string coding, the schema theorem and the formation of building blocks in the strings. We consider the coding of continuous-valued variables and bit array representations, elitism, methods of maintaining diversity in the population and introduce a further illustration in structural optimization. The application of the genetic algorithm is then extended into large scale-situations, particularly design situations involving a large number of variables. A combinatorial space reduction heuristic based on a record of parameter selection intensities is described. The allocation of fitness to partial strings is reviewed. Consideration is given to the multi-objective GA and pareto optimality. There follows a brief introduction to mathematical models of the GA. The GA is used to train a neural network as an alternative to back-propagation. We consider the ‘permutations’ problem and introduce the concept of ‘shift’. The method is illustrated by training a neural network for structural analysis. The chapter concludes with a brief review of the implicit parallelism of the GA and suggestions as to how the algorithm might be improved with parallel hardware and a further example application.
••
06 Jul 1999TL;DR: The present paper proposes some design guidelines for crossover operators for RCGA, and a multi-parental extension of the UNDX is proposed so as to enhance its exploration ability.
Abstract: The unimodal normal distribution crossover (UNDX) for the real-coded genetic algorithms (RCGA) proposed by Ono et al. (1997, 1998) shows an excellent performance in optimization problems of multi-modal and highly epistatic fitness functions in continuous search space. Further, theoretical analysis of the UNDX shows that the UNDX is a crossover operator that preserves the statistics such as the mean vector and the covariance matrix of the population well. The present paper proposes some design guidelines for crossover operators for RCGA. Then, based on these guidelines, a multi-parental extension of the UNDX is proposed so as to enhance its exploration ability. Performance of the extended UNDX is evaluated by numerical experiments.
••
TL;DR: A genetic algorithm based on random keys representation, elitist reproduction, Bernoulli crossover and immigration type mutation is developed and Convergence of the algorithm is proved.
Abstract: This paper considers the scheduling problem to minimize total tardiness given multiple machines, ready times, sequence dependent setups, machine downtime and scarce tools. We develop a genetic algorithm based on random keys representation, elitist reproduction, Bernoulli crossover and immigration type mutation. Convergence of the algorithm is proved. We present computational results on data sets from the auto industry. To demonstrate robustness of the approach, problems from the literature of different structure are solved by essentially the same algorithm.
••
TL;DR: A genetic algorithm (GA) approach to support interactive planning of a piping route path in plant layout design is presented and some simulation results using the prototype system are presented to show the validity of this approach.
Abstract: A genetic algorithm (GA) approach to support interactive planning of a piping route path in plant layout design is presented. To present this approach, the paper mainly describes the basic ideas used in the methodology, which include the definition of genes to deal with pipe routes, the concept of spatial potential energy, the method of generating initial individuals for GA optimization, the zone concept in route generation using GAs, the evaluation of crossover methods, and definition and application of fitness functions. The prototype system that has been developed based on the methodology gives designers an environment to design a piping route path in an interactive and collaborative manner with a very simple operation. The GA optimization technique generates a route path through evolution of genes that represent the pipe route. A designer evaluates the route path, modifies it, conducts another GA optimization and/or repeats the procedure until the appropriate route is desig ned. The paper also presents some simulation results using the prototype system to show the validity of this approach.
•
01 May 1999
TL;DR: This book provides comprehensive coverage of the techniques involved, describing the intrinsic characteristics, advantages and constraints of genetic algorithms, as well as discussing genetic operations such as crossover, mutation and reinsertion.
Abstract: From the Publisher:
The practical application of Genetic Algorithms to the solution of engineering problems is rapidly becoming an established approach in the fields of control and signal processing. This book provides comprehensive coverage of the techniques involved, describing the intrinsic characteristics, advantages and constraints of genetic algorithms, as well as discussing genetic operations such as crossover, mutation and reinsertion. In addition, the principle of multiobjective optimization and computing parallelism are discussed. These features are fully illustrated by real-world applications. Also described is a newly proposed and unique hierarchical genetic algorithm designed to address the problems in determining system topology.
•
13 Jul 1999TL;DR: The genetic algorithms for the unconstrained binary quadratic programming problem (BQP) are presented and it is shown that for small problems a simple genetic algorithm with uniform crossover is sufficient to find optimum or best-known solutions in short time, while for problems with a high number of variables it is essential to incorporate local search to arrive at high-quality solutions.
Abstract: In this paper, genetic algorithms for the unconstrained binary quadratic programming problem (BQP) are presented. It is shown that for small problems a simple genetic algorithm with uniform crossover is sufficient to find optimum or best-known solutions in short time, while for problems with a high number of variables (n ≥ 200) it is essential to incorporate local search to arrive at high-quality solutions. A hybrid genetic algorithm incorporating local search is tested on 40 problem instances of sizes containing between n = 200 and n = 2500. The results of the computer experiments show that the approach is comparable to alternative heuristics such as tabu search for small instances and superior to tabu search and simulated annealing for large instances. New best solutions could be found for 14 large problem instances.
••
TL;DR: In this paper, the application of genetic algorithms as a global search technique for a quick identification of optimal or near optimal operation sequences in a dynamic planning environment is presented, where a novel initialization scheme for representing the genetic code and a new crossover operator are designed to retain the local operation precedence for each operation precedence.
Abstract: Computer aided process planning (CAPP) is an important interface between computer aided design (CAD) and computer aided manufacturing (CAM) in computer integrated manufacturing (CIM). Operation sequencing in process planning is concerned with the selection of machining operations in steps that can produce each form feature of the part by satisfying relevant technological constraints specified in the part drawing. A single sequence of operations may not be the best for all the situations in a changing production environment with multiple objectives such as minimizing number of set-ups, maximizing machine utilization and minimizing number of tool changes. This paper demonstrates the application of genetic algorithms as a global search technique for a quick identification of optimal or near optimal operation sequences in a dynamic planning environment. A novel initialization scheme for representing the genetic code and a new crossover operator are designed to retain the local operation precedence for each fo...
••
TL;DR: A Genetic Algorithm (GA) based system, coupling the computer codes GENESIS 5.0 and ANC through the interface ALGER has been developed aiming at pressurized water reactor's (PWR) fuel management optimization, with an innovative codification, the List Model (LM), incorporated into the system.
••
TL;DR: A new search strategy, called systematic crossover, couples the best individuals, tests every possible crossover point, and takes the two best individuals for the next generation, which is significantly faster in identifying the global minimum than the standard genetic algorithm.
Abstract: To improve protein folding simulations, we investigated a new search strategy in combination with the simple genetic algorithm on a two-dimensional lattice model. This search strategy, we called systematic crossover, couples the best individuals, tests every possible crossover point, and takes the two best individuals for the next generation. We compared the standard genetic algorithm with and without this new implementation for various chain lengths and showed that this strategy finds local minima with better energy values and is significantly faster in identifying the global minimum than the standard genetic algorithm.
••
TL;DR: A simple random selection strategy that implements iterative generation of models, but directly avoids crossover and mutation, has been developed and is implemented herein to rapidly identify from a pool of allowable variables those which are most closely associated with a given response variable.
Abstract: Variable selection is typically a time-consuming and ambiguous procedure in performing quantitative structure−activity relationship (QSAR) studies on overdetermined (regressor-heavy) data sets. A variety of techniques including stepwise and partial least squares/principlal components analysis (PLS/PCA) regression have been applied to this common problem. Other strategies, such as neural networks, cluster significance analysis, nearest neighbor, or genetic (function) or evolutionary algorithms have also evaluated. A simple random selection strategy that implements iterative generation of models, but directly avoids crossover and mutation, has been developed and is implemented herein to rapidly identify from a pool of allowable variables those which are most closely associated with a given response variable. The FRED (fast random elimination of descriptors) algorithm begins with a population of offspring models composed of either a fixed or variable number of randomly selected variables. Iterative eliminati...
••
TL;DR: The modified GA is tested on the New York City water supply expansion problem and obtains the lowest‐cost feasible solution reported in the literature in far fewer generations than any previous GA.
Abstract: A modified genetic algorithm (GA) is proposed for water distribution network optimization. Several changes are introduced in the selection and mutation processes of a simple GA. In each generation a constant number of solutions is eliminated, the selected ones are ranked for crossover, and the new solutions are allowed to undergo at most one mutation. All these modifications greatly increase the algorithm convergence. The modified GA is tested on the New York City water supply expansion problem. It obtains the lowest-cost feasible solution reported in the literature in far fewer generations than any previous GA.
••
10 May 1999TL;DR: The hybrid genetic algorithms can produce near-optimal solutions for problems of sizes up to 25 percent bigger than what can be solved previously by utilizing optimal but possibly partial solutions from dynamic programming.
Abstract: This paper presents a novel approach to solving the single-vehicle pickup and delivery problem with time windows and capacity constraints. While dynamic programming has been used to find the optimal routing to a given problem, it requires time exponential in the number of tasks. Therefore, it often fails to find the solutions under real-time conditions in an automated factory. This research explores anytime problem solving using genetic algorithms. By utilizing optimal but possibly partial solutions from dynamic programming, the hybrid genetic algorithms can produce near-optimal solutions for problems of sizes up to 25 percent bigger than what can be solved previously. This paper reports the experimental results of the proposed hybrid approach with four different crossover operators as well as three mutation operators. The experiments demonstrated the advantages of the hybrid approach with respect to dynamic task requests.
••
16 Jul 1999
TL;DR: This paper presents an example of an evolutionary algorithm using crossover and shows that it is essentially more efficient than evolutionary algorithms without crossover.
Abstract: There is a lot of experimental evidence that crossover is, for some functions, an essential operator of evolutionary algorithms. Nevertheless, it was an open problem to prove for some function that an evolutionary algorithm using crossover is essentially more efficient than evolutionary algorithms without crossover. In this paper, such an example is presented and its properties are proved.
••
TL;DR: Evidence suggesting that crossover alone, operating on graphs, can evolve any possible molecule given an appropriate fitness function and a population containing both rings and chains is presented.
Abstract: Molecular nanotechnology is the precise, three-dimensional control of materials and devices at the atomic scale. An important part of nanotechnology is the design of molecules for specific purposes. This paper describes early results using genetic software techniques to automatically design molecules under the control of a fitness function. The fitness function must be capable of determining which of two arbitrary molecules is better for a specific task. The software begins by generating a population of random molecules. The individual molecules in a population are then evolved towards greater fitness by randomly combining parts of the better existing molecules to create new molecules. These new molecules then replace some of the less fit molecules in the population. We apply a unique genetic crossover operator to molecules represented by graphs, i.e., sets of atoms and the bonds that connect them. We present evidence suggesting that crossover alone, operating on graphs, can evolve any possible molecule given an appropriate fitness function and a population containing both rings and chains. Most prior work evolved strings or trees that were subsequently processed to generate molecular graphs. In principle, genetic graph software should be able to evolve other graph-representable systems such as circuits, transportation networks, metabolic pathways, and computer networks.
••
TL;DR: A detailed illustrative example is presented to demonstrate that GA is capable of finding global or near-global optimum solutions of multi-modal functions.
Abstract: In this paper, an attractive approach for teaching genetic algorithm (GA) is presented. This approach is based primarily on using MATLAB in implementing the genetic operators: crossover, mutation and selection. A detailed illustrative example is presented to demonstrate that GA is capable of finding global or near-global optimum solutions of multi-modal functions. An application of GA in designing a robust controller for uncertain control systems is also given to show its potential in designing engineering intelligent systems.