## Real-coded Memetic Algorithms with crossover hill-climbing (2004)

### Cached

### Download Links

- [sci2s.ugr.es]
- [sci2s.ugr.es]
- [sci2s.ugr.es]
- DBLP

### Other Repositories/Bibliography

Venue: | Evolutionary Computation |

Citations: | 35 - 8 self |

### BibTeX

@ARTICLE{Lozano04real-codedmemetic,

author = {Manuel Lozano and Francisco Herrera and Natalio Krasnogor and Daniel Molina},

title = {Real-coded Memetic Algorithms with crossover hill-climbing},

journal = {Evolutionary Computation},

year = {2004},

pages = {3--273}

}

### OpenURL

### Abstract

This paper presents a real-coded memetic algorithm that applies a crossover hillclimbing to solutions produced by the genetic operators. On the one hand, the memetic algorithm provides global search (reliability) by means of the promotion of high levels of population diversity. On the other, the crossover hill-climbing exploits the selfadaptive capacity of real-parameter crossover operators with the aim of producing an effective local tuning on the solutions (accuracy). An important aspect of the memetic algorithm proposed is that it adaptively assigns different local search probabilities to individuals. It was observed that the algorithm adjusts the global/local search balance according to the particularities of each problem instance. Experimental results show that, for a wide range of problems, the method we propose here consistently outperforms other real-coded memetic algorithms which appeared in the literature.

### Citations

1164 | Hundbook of Genetic Algorithms
- Davis, ed
- 1991
(Show Context)
Citation Context ...ished that pure genetic algorithms (GAs) are not well suited to fine tuning search in complex search spaces, and that hybridisation with other techniques can greatly improve the efficiency of search (=-=Davis, 1991-=-; Goldberg and Voessner, 1999). GAs that have been hybridized with local search techniques (LS) are often called memetic algorithms (MAs) (Moscato, 1989; Moscato, 1999). MAs are evolutionary algorithm... |

1065 |
Multi-Objective Optimization Using Evolutionary Algorithms
- Deb
- 2001
(Show Context)
Citation Context ...tor of floating point numbers whose size is kept the same as the length of the vector, which is the solution to the problem. GAs based on real number representation are called real-coded GAs (RCGAs) (=-=Deb, 2001-=-; Herrera, Lozano and Verdegay, 1998). For function optimisation problems in continuous search spaces, an important difficulty must be addressed: solutions of high precision must be obtained by the so... |

858 | An Analysis of the Behavior of a Class of Genetic Adaptive Systems. (Doctoral Dissertation - DeJong - 1975 |

661 | Evolutionsstrategie: Optimierung technischer Systeme nach Prinzipien der biologischen Evolution. Frommann-Holzboog - Rechenberg - 1973 |

510 |
Numerical Optimization of Computer Models
- Schwefel
- 1981
(Show Context)
Citation Context ...best of their offspring. Most well-known continuous local searchers (derivative-free), such as the Solis and Wets’ algorithm (Solis and Wets, 1981) and the (1+1)-evolution strategy (Rechenberg, 1973; =-=Schwefel, 1981-=-), make use of explicit control parameters (e.g., step sizes) to guide the search. In addition, they adapt the parameters, in such a way that the moves being made may be of varying sizes, depending on... |

391 | A comparative analysis of selection schemes used in genetic algorithms. Foundations of Genetic Algorithms - Goldberg, Deb - 1991 |

344 | Predictive models for the breeder genetic algorithm i. continuous parameter optimization - Mühlenbein, Schlierkamp-Voosen - 1993 |

277 | Differential evolution: a simple and efficient adaptive scheme for global optimization over continuous spaces
- Storn, Price
- 1995
(Show Context)
Citation Context ...omains: Systems of Linear Equations (Eshelman, Mathias and Schaeffer, 1997), Frequency Modulation Sounds Parameter Identification Problem (Tsutsui and Fujimoto, 1993), and Polynomial Fitting Problem (=-=Storn and Price, 1995-=-). They are described below. Systems of Linear Equations The problem may be stated as solving for the elements of a vector, �, given the matrix � and vector � in the expression: � � � �. The evaluatio... |

259 |
The CHC Adaptive Search Algorithm, How to Have Safe Search When Engaging
- Eshelman
(Show Context)
Citation Context ...echanism determines the way the chromosomes are mated for applying the crossover to them (Step 1 in Figure 2). Mates can be selected so as to favour population diversity (Craighurst and Martin, 1995; =-=Eshelman, 1991-=-; Fernandes and Rosa, 2001). A way to do this is the negative assortative mating mechanism. Assortative mating is the natural occurrence of mating between individuals of similar phenotype more or less... |

188 | The parallel genetic algorithm as function optimizer - Mühlenbein, Schomisch, et al. - 1991 |

187 | On evolution, search, optimization, genetic algorithms and martial arts, towards memetic algorithms
- Moscato
- 1989
(Show Context)
Citation Context ...ues can greatly improve the efficiency of search (Davis, 1991; Goldberg and Voessner, 1999). GAs that have been hybridized with local search techniques (LS) are often called memetic algorithms (MAs) (=-=Moscato, 1989-=-; Moscato, 1999). MAs are evolutionary algorithms that apply a separate LS process to refine individuals (e.g. improve their fitness by hill-climbing). An important aspect concerning MAs is the trade-... |

138 |
Real-coded genetic algorithms and interval-schemata,” in Foundations of Genetic Algorithms-2
- Eshelman, Schaffer
- 1993
(Show Context)
Citation Context ...er crossover operators use the variation of the population to constrain the search and bias the distribution of offspring (i.e., sampling bias). Some real-parameter crossover operators, such as BLX- (=-=Eshelman and Schaffer, 1993-=-), have a sampling bias that favour the generation of offspring being gathered towards the center of the region where the parental population covers. In other words, they tend to search the interpolat... |

127 | Tackling real-coded genetic algorithms: Operators and tools for behavioural analysis - Herrera, Lozano, et al. - 1998 |

114 | Self-Adaptation in Genetic Algorithms
- Bäck
- 1992
(Show Context)
Citation Context ...ns: Sphere model (� �) (De Jong, 1975; Schwefel, 1981), Generalized Rosenbrock’s function (� ) (De Jong, 1975), Schwefel’s Problem 1.2 (� �) (Schwefel, 1981), Generalized Rastringin’s function (� � )(=-=Bäck, 1992-=-; Törn and Antanas, 1989), Griewangk’s function (�� �) (Griewangk, 1981). Figure 5 shows their formulation. The dimension of the search space is 25. � � is a continuous, strictly convex, and unimodal ... |

113 |
Minimization by random search techniques
- Solis, Wets
- 1981
(Show Context)
Citation Context ...stood as promoting incest between the best of the current parents and the best of their offspring. Most well-known continuous local searchers (derivative-free), such as the Solis and Wets’ algorithm (=-=Solis and Wets, 1981-=-) and the (1+1)-evolution strategy (Rechenberg, 1973; Schwefel, 1981), make use of explicit control parameters (e.g., step sizes) to guide the search. In addition, they adapt the parameters, in such a... |

103 |
Memetic algorithms: A short introduction, in
- Moscato
- 1999
(Show Context)
Citation Context ... improve the efficiency of search (Davis, 1991; Goldberg and Voessner, 1999). GAs that have been hybridized with local search techniques (LS) are often called memetic algorithms (MAs) (Moscato, 1989; =-=Moscato, 1999-=-). MAs are evolutionary algorithms that apply a separate LS process to refine individuals (e.g. improve their fitness by hill-climbing). An important aspect concerning MAs is the trade-off between the... |

86 |
Preventing premature convergence in genetic algorithms by preventing incest
- Eshelman, Schaffer
- 1991
(Show Context)
Citation Context ...lize Standard-replacement-strategy to insert � in population. 13. Employ Standard-replacement-strategyto insert � in population. Figure 3: Pseudocode algorithm for the Steady-State RCMA proposed GAs (=-=Eshelman and Schaffer, 1991-=-). In the MA literature, keeping population diversity while using LS together with a GA is always an issue to be addressed, either implicitly or explicitly (Krasnogor, 2002). We review some approaches... |

86 | Adaptive global optimization with local search
- Hart
- 1994
(Show Context)
Citation Context ...mbination of RCGAs with some type of LS mechanism is denoted RCMAs. They are motivated by the apparent need to employ both a global and LS strategy to provide an effective global optimisation method (=-=Hart, 1994-=-). RCMA instances presented in the literature may be classified into two different groups: Hybrid RCGAs. They use efficient LIPs on continuous domains, e.g., hill-climbers for nonlinear optimisation (... |

86 | An Overview of Evolutionary Algorithms - Whitley - 2001 |

76 | Using selection to improve particle swarm optimization - Angeline - 1998 |

70 | Crossover or mutation
- Spears
- 1992
(Show Context)
Citation Context ...proposed, which combines a disruptive crossover operator with a conservative selection strategy. 5.3.2 BGA Mutation Operator The mutation operator serves to create random diversity in the population (=-=Spears, 1993-=-). In the case of working with real coding, a topic of major importance involves the control of the proportion or strength in which real-coded genes are mutated, i.e., the step size (Bäck, 1996). Diff... |

66 | Real-coded genetic algorithms with simulated binary crossover: Studies on multi-modal and multi-objective problems
- Deb, Kumar
- 1995
(Show Context)
Citation Context ...the recent past, RCGAs with some of these crossovers have been demonstrated to exhibit self-adaptive behavior similar to that observed in evolution strategies and evolutionary programming approaches (=-=Deb and Beyer, 2001-=-; Kita, 2001). Moreover, Beyer and Deb (2001) argue that a variation operator that harness the difference of the parents in the search space is essential for the resulting evolutionary algorithm to ex... |

61 | A formal analysis of the role of multi-point crossover in genetic algorithms - Jong, Spears - 1992 |

61 | Adapting crossover in evolutionary algorithms
- Spears
- 1995
(Show Context)
Citation Context ...ps to locate more promising areas in the search space regions being refined. 6.4 Study of the Adaptive Mechanism There are, at least, two ways to study the operation of an adaptive mechanism for GAs (=-=Spears, 1995-=-). The first is from the point of view of performance (test functions are commonly used to evaluate performance improvement). The second view is quite different in that it ignores performance and conc... |

56 | A computationally efficient evolutionary algorithm for real-parameter optimization - DEB, Anand, et al. |

54 | Edge assembly crossover: A high-power genetic algorithm for the travelling salesman problem - Nagata, Kobayashi - 1997 |

49 |
Studies on the theory and design space of memetic algorithms
- Krasnogor
- 2002
(Show Context)
Citation Context ...A proposed GAs (Eshelman and Schaffer, 1991). In the MA literature, keeping population diversity while using LS together with a GA is always an issue to be addressed, either implicitly or explicitly (=-=Krasnogor, 2002-=-). We review some approaches: Mühlenbein, Schomisch and Born (1991) integrate LS procedures to distributed GAs, which keep, in parallel, several sub-populations that are processed by genetic algorithm... |

47 | Memetic algorithms for combinatorial optimization: fitness landscapes and effective search strategies - Merz - 1998 |

47 | Strategy adaptation by competing subpopulations
- Schlierkamp-Voosen, Mühlenbein
(Show Context)
Citation Context ...ce they must continually change their search direction to reach the optimum. This function has been considered by some authors to be a real challenge for any continuous function optimization program (=-=Schlierkamp-Voosen and Mühlenbein, 1994-=-). A great part of its difficulty lies in the fact that there are nonlinear interactions between the variables, i.e., it is nonseparable (Whitley, Rana, Dzubera and Mathias). � � is a continuous and u... |

44 | Gradual distributed real-coded genetic algorithms
- Herrera, Lozano
(Show Context)
Citation Context ...of the proportion or strength in which real-coded genes are mutated, i.e., the step size (Bäck, 1996). Different techniques have been suggested for the control of the step size during the RCGA’s run (=-=Herrera and Lozano, 2000-=-a; Smith and Fogarty, 1997). An example is the non-uniform mutation, which is considered to be one of the most suitable mutation operators for RCGAs (Herrera, Lozano, Verdegay, 1998). Its main idea is... |

44 |
Operator and parameter adaptation in genetic algorithms
- Smith, Fogarty
- 1997
(Show Context)
Citation Context ...th in which real-coded genes are mutated, i.e., the step size (Bäck, 1996). Different techniques have been suggested for the control of the step size during the RCGA’s run (Herrera and Lozano, 2000a; =-=Smith and Fogarty, 1997-=-). An example is the non-uniform mutation, which is considered to be one of the most suitable mutation operators for RCGAs (Herrera, Lozano, Verdegay, 1998). Its main idea is to decrease the step size... |

41 | S.P.: Hybrid methods using genetic algorithms for global optimization - Renders, Flasse - 1998 |

39 | Optimizing global-local search hybrids
- Goldberg, Voessner
- 1999
(Show Context)
Citation Context ...re genetic algorithms (GAs) are not well suited to fine tuning search in complex search spaces, and that hybridisation with other techniques can greatly improve the efficiency of search (Davis, 1991; =-=Goldberg and Voessner, 1999-=-). GAs that have been hybridized with local search techniques (LS) are often called memetic algorithms (MAs) (Moscato, 1989; Moscato, 1999). MAs are evolutionary algorithms that apply a separate LS pr... |

35 | Minimal Generation Gap Model for GAs considering both Exploration and Exploitation - Satoh, Yamamura, et al. - 1996 |

34 | Hybridizing genetic algorithms with hill-climbing methods for global optimization: Two possible ways - Renders, Bersini - 1994 |

34 |
Forking genetic algorithm with blocking and shrinking modes (FGA
- Tsutsui, Fujimoto
- 1993
(Show Context)
Citation Context ...ization problems of parameters with variables on continuous domains: Systems of Linear Equations (Eshelman, Mathias and Schaeffer, 1997), Frequency Modulation Sounds Parameter Identification Problem (=-=Tsutsui and Fujimoto, 1993-=-), and Polynomial Fitting Problem (Storn and Price, 1995). They are described below. Systems of Linear Equations The problem may be stated as solving for the elements of a vector, �, given the matrix ... |

34 | Multi-parent Re- combination with Simplex Crossover in Real Coded Genetic Algorithms - Tsutsui, Yamamura, et al. - 1999 |

32 | On Self-Adaptive Features in Real-Parameter Evolutionary Algorithms
- Beyer, Deb
(Show Context)
Citation Context ...since there are real-parameter crossover operators having a selfadaptive nature in that they can generate offspring adaptively according to the distribution of parents without any adaptive parameter (=-=Beyer and Deb, 2001-=-, Kita, 2000). With the passing of generations, the RCMA loses diversity, which allows the crossover to create offspring distributed densely around the parents, inducing an effective local tuning. Thi... |

31 | A memetic algorithm with self-adaptive local search: TSP as a case study - Krasnogor, Smith - 2000 |

30 | An empirical investigation of multiparent recombination operators in evolution strategies - Eiben, Back - 1997 |

28 |
Evolutionary algorithms with local search for combinatorical optimization
- Land
- 1998
(Show Context)
Citation Context ...obability, , is introduced, which determines the probability that LS will be invoked to refine a new chromosome. Steady-state MAs integrate global and local search more tightly than generational MAs (=-=Land, 1998-=-). This interleaving of the global and local search phases allows the two to influence each other, e.g., the SSGA chooses good starting points, and LS provides an accurate representation of that regio... |

26 | When Selection Meets Seduction - Ronald - 1993 |

25 |
Evolutionary algorithms and gradient search: similarities and differences
- Salomon
- 1998
(Show Context)
Citation Context ...orithms, which are designed using the foundational ideas of LIPs. Two examples are the evolutionary pattern search algorithm (Hart, 2001a; Hart, 2001b) and the evolutionary gradient search procedure (=-=Salomon, 1998-=-). RCMAs with crossover-based LS algorithms. The crossover operator is a recombination operator that produces elements around the parents. For that reason, it may be considered to be a move operator f... |

24 | Tuning evolutionary programming for conformationally flexible molecular docking - Gehlhaar, Fogel - 1996 |

24 | A taxonomy for the crossover operator for real-coded genetic algorithms: an experimental study - Herrera, Lozano, et al. |

23 |
Generalized Descent of Global Optimization
- Griewangk
- 1981
(Show Context)
Citation Context ...d Rosenbrock’s function (� ) (De Jong, 1975), Schwefel’s Problem 1.2 (� �) (Schwefel, 1981), Generalized Rastringin’s function (� � )(Bäck, 1992; Törn and Antanas, 1989), Griewangk’s function (�� �) (=-=Griewangk, 1981-=-). Figure 5 shows their formulation. The dimension of the search space is 25. � � is a continuous, strictly convex, and unimodal function. � is a continuous and unimodal function, with the optimum loc... |

21 | Crossover, macromutation, and population-based search - Jones - 1995 |

19 | Hybridized crossover-based search techniques for program discovery - O’Reilly, Oppacher - 1995 |

18 | Empirical investigation of the benefits of partial Lamarckianism - Houck, Joines, et al. - 1997 |

17 |
Enhancing GA performance through crossover prohibitions based on ancestry
- Craighurst, Martin
- 1995
(Show Context)
Citation Context ... Mating The mating selection mechanism determines the way the chromosomes are mated for applying the crossover to them (Step 1 in Figure 2). Mates can be selected so as to favor population diversity (=-=Craighurst and Martin, 1995-=-; Eshelman, 1991; Fernandes and Rosa, 2001). A way to do this is the negative assortative mating mechanism. Assortative mating is the natural occurrence of mating between individuals of similar phenot... |