## Informed operators: Speeding up genetic-algorithm-based design optimization using reduced models (2000)

Venue: | In Proceedings of the Genetic and Evolutionary Computation Conference (GECCO |

Citations: | 17 - 8 self |

### BibTeX

@INPROCEEDINGS{Rasheed00informedoperators:,

author = {Khaled Rasheed and Haym Hirsh},

title = {Informed operators: Speeding up genetic-algorithm-based design optimization using reduced models},

booktitle = {In Proceedings of the Genetic and Evolutionary Computation Conference (GECCO},

year = {2000},

pages = {628--635},

publisher = {Morgan Kaufmann}

}

### Years of Citing Articles

### OpenURL

### Abstract

In this paper we describe a method for improving genetic-algorithm-based optimization using informed genetic operators. The idea is to make the genetic operators such as mutation and crossover more informed using reduced models. In every place where a random choice is made, for example when a point is mutated, instead of generating just one random mutation we generate several, rank them using a reduced model, then take the best to be the result of the mutation. The proposed method is particularly suitable for search spaces with expensive evaluation functions, such as arise in engineering design. Empirical results in several engineering design domains demonstrate that the proposed method can significantly speed up the GA optimizer. 1 Introduction This paper concerns the application of Genetic Algorithms (GAs) in realistic engineering design domains. In such domains a design is represented by a number of continuous design parameters, so that potential solutions are vec...

### Citations

86 |
Using genetic algorithms in engineering design optimization with nonlinear constraints
- Powell, Skolnick
- 1993
(Show Context)
Citation Context ... and compared their performance. Those problems have become used in engineering design optimization domains as benchmarks. One of the recent experiments involving these domains was reported by Powell =-=[6]-=-, in which a GA package called OOGA and a numerical optimization package called NumOpt were compared to each other in 10 of Sandgren 's domains. The 10 domains were a representative sample of the orig... |

28 | GADO: A genetic algorithm for continuous design optimization
- Rasheed
- 1998
(Show Context)
Citation Context ...odel to be a better-than-random approximation of the accurate one. Our approach is also simple and easy to add to an existing genetic algorithm. We conducted our investigations in the context of GADO =-=[7, 10]-=-, a GA that was designed with the goal of being suitable for use in engineering design. It uses new operators and search control strategies that target the domains that typically arise in such applica... |

22 | Using modeling knowledge to guide design space search
- Gelsey, Schwabacher, et al.
- 1996
(Show Context)
Citation Context ...ircraft design domain 3.1.1 Domain description Our first domain concerns the conceptual design of supersonic transport aircraft. We summarize it briefly here; it is described in more detail elsewhere =-=[4]-=-. Figure 1 shows a diagram of a typical airplane automatically designed by our software system. The GA attempts to find a good design for a particular mission by varying twelve of the aircraft concept... |

14 | Learning to be selective in genetic-algorithm-based design optimization
- Rasheed, Hirsh
- 1999
(Show Context)
Citation Context ...odel to be a better-than-random approximation of the accurate one. Our approach is also simple and easy to add to an existing genetic algorithm. We conducted our investigations in the context of GADO =-=[7, 10]-=-, a GA that was designed with the goal of being suitable for use in engineering design. It uses new operators and search control strategies that target the domains that typically arise in such applica... |

14 |
The utility of nonlinear programming algorithms
- Sandgren
- 1977
(Show Context)
Citation Context ... other parameters intact. We then compared the two systems in several domains: two domains from real tasks in aerodynamic design, plus two others from an existing set of engineering design benchmarks =-=[11]-=-. 0 17.5813 45.0132 62.5945 0 1.6764 5.53123 -12.8733 0 12.8733 engineScale=1.45576 wing_area(m^2)=420.307 wing_aspect_ratio=1.57715 fuselage_taper_length(m)=35.1625 wing_t_over_c=0.0301626 fuel_annul... |

14 | High performance supersonic missile inlet design using automated optimization
- Zha, Smith, et al.
- 1996
(Show Context)
Citation Context ...le inlet domain 3.2.1 Domain description Our second domain concerns the design of inlets for supersonic and hypersonic missiles. We summarize it briefly here; it is described in more detail elsewhere =-=[14]-=-. The missile inlet designed is an axisymmetric mixed compression inlet that cruises at Mach 4 at 60,000 feet altitude. Minimum manufacture cost for this inlet is critical, and therefore, techniques s... |

13 |
Design by natural selection
- Dunham, Fridshal, et al.
- 1963
(Show Context)
Citation Context ...e it takes to run a simulator, this policy could result in excellent speed-up. The use of reduced models to save time in evolutionary optimization dates all the way back to the sixties. Dunham et al. =-=[1]-=- worked with a two level problem in which they used an approximate model most of the time and only used the accurate/expensive model in the final stages of refinement. Numerous research e#orts compute... |

9 | An incremental-approximate-clustering approach for developing dynamic reduced models for design optimization
- Rasheed
- 2002
(Show Context)
Citation Context ...nts described below, we used reduced models that are acquired on-line as the optimization progresses. The technique for computing these dynamic reduced models is described in further detail elsewhere =-=[9]-=-. The idea is to maintain a large sample of the points encountered in the course of the optimization divided into dynamic clusters and periodically compute least squares approximations for the measure... |

8 |
Parmee, “The use of a co-operative multi-level CHC GA for structural shape optimization
- Vekeria, C
- 1996
(Show Context)
Citation Context ... with no looking back [12]. Other approaches rely on special relations between the approximate and accurate model to develop interesting multi-level search strategies. A notable class of such methods =-=[13, 2]-=- focus on building variants of injection island genetic algorithms (iiGAs) for problems involving finite element analysis models. The approach was to have many islands using low accuracy /cheap evalua... |

5 | Guided crossover: A new operator for genetic algorithm based optimization - Rasheed - 1999 |

4 |
Evaluation of injection island GA performance pn flywheel design optimization
- Eby, Averill, et al.
- 1998
(Show Context)
Citation Context ... with no looking back [12]. Other approaches rely on special relations between the approximate and accurate model to develop interesting multi-level search strategies. A notable class of such methods =-=[13, 2]-=- focus on building variants of injection island genetic algorithms (iiGAs) for problems involving finite element analysis models. The approach was to have many islands using low accuracy /cheap evalua... |

3 |
Application of genetic programming to the choice of a structure of global approximations
- Toropov, Alvarez
- 1998
(Show Context)
Citation Context .../expensive model in the final stages of refinement. Numerous research e#orts compute a response surface approximation and use it instead of the very expensive evaluation function with no looking back =-=[12]-=-. Other approaches rely on special relations between the approximate and accurate model to develop interesting multi-level search strategies. A notable class of such methods [13, 2] focus on building ... |

1 | Topographical mapping assisted evolutionary search for multilevel optimization
- El-Beltagy, Keane
- 1999
(Show Context)
Citation Context ... using low accuracy /cheap evaluation models with small numbers of finite elements that progressively propagate individuals to fewer islands using more accurate/expensive evaluations. Some approaches =-=[3]-=- use both an accurate /expensive and a cheaper/approximate model interchangeably during the search by relying on intelligent techniques to decide when it is safe to rely on the cheaper model. We obser... |