Results 1  10
of
10
Asynchronous Evolutionary MultiObjective Algorithms with Heterogeneous Evaluation Costs
, 2011
"... (EAs) is straightforward, by distributing all fitness computations to slaves. The benefits of asynchronous steadystate approaches are wellknown when facing a possible heterogeneity among the evaluation costs in term of runtime, be they due to heterogeneous hardware or nonlinear numerical simulatio ..."
Abstract

Cited by 1 (1 self)
 Add to MetaCart
(Show Context)
(EAs) is straightforward, by distributing all fitness computations to slaves. The benefits of asynchronous steadystate approaches are wellknown when facing a possible heterogeneity among the evaluation costs in term of runtime, be they due to heterogeneous hardware or nonlinear numerical simulations. However, when this heterogeneity depends on some characteristics of the individuals being evaluated, the search might be biased, and some regions of the search space poorly explored. Motivated by a realworld case study of multiobjective optimization problem – the optimization of the combustion in a Diesel Engine – the consequences of different components of heterogeneity in the evaluation costs on the convergence of two Evolutionary Multiobjective Optimization Algorithms are investigated on artificiallyheterogeneous benchmark problems. In some cases, better spread of the population on the Pareto front seem to result from the interplay between the heterogeneity at hand and the evolutionary search. I.
de l’Université de Rouen (Spécialité Génie Informatique, Automatique et Traitement du Signal) Soutenue le 29/11/2011 Composition du jury
, 2012
"... Tombre d’avoir accepté d’être les rapporteurs de ce document. Ils sont des références pour moi et j’ai beaucoup apprécié leur travail d’expertise. Je remercie aussi JeanMarc Ogier d’avoir accepté mon invitation et d’avoir présidé ce jury. JeanMarc est la personne qui m’a donné le goût de la recher ..."
Abstract
 Add to MetaCart
(Show Context)
Tombre d’avoir accepté d’être les rapporteurs de ce document. Ils sont des références pour moi et j’ai beaucoup apprécié leur travail d’expertise. Je remercie aussi JeanMarc Ogier d’avoir accepté mon invitation et d’avoir présidé ce jury. JeanMarc est la personne qui m’a donné le goût de la recherche et ses qualités humaines et scientifiques sont trop nombreuses pour les lister ici. Mention spéciale aux collègues locaux de ce jury. Yves et Laurent ont pris le relai de JeanMarc quand ce dernier est parti chercher ses fameuses 2250 heures de soleil par an sur la côte atlantique. J’apprécie énormément de travailler avec eux, et j’espère que ce n’est qu’un début. tel00671168, version 1 8 Oct 2012 Je remercie aussi vivement les nombreux doctorants et stagiaires avec qui j’ai travaillé ces dix dernières années. Les encadrer a été un véritable plaisir et je leur dois pour beaucoup les résultats obtenus. Coté laboratoire, là encore les personnes auxquelles je voudrais témoigner ma reconnaissance sont très nombreuses. Je pense que travailler au LITIS est une chance, pour l’ambiance et la qualité des travaux qui y sont menés. Parmi tous les collègues, une mention particulière va à Pierrot et Clem. Ce sont mes binômes de travail et des amis, et j’espère qu’on va avoir l’occasion de travailler encore beaucoup ensemble. Merci également à Thierry avec qui c’est un réel plaisir de travailler. Une spéciale dédicace aussi à super Fabienne dont l’efficacité est impressionnante.
Zhiyang Onga,c, Andy HaoWei Loa, Matthew Berrymana,b, and Derek Abbotta,b aSchool of Electrical and Electronic Engineering, and bCentre for Biomedical Engineering,
"... tradeoff between pleiotropy and redundancy ..."
Approximation of Digital Curves using a MultiObjective Genetic Algorithm
"... In this paper, a digital planar curve approximation method based on a multiobjective genetic algorithm is proposed. In this method, the optimization/exploration algorithm locates breakpoints on the digital curve by minimizing simultaneously the number of breakpoints and the approximation error. Usi ..."
Abstract
 Add to MetaCart
(Show Context)
In this paper, a digital planar curve approximation method based on a multiobjective genetic algorithm is proposed. In this method, the optimization/exploration algorithm locates breakpoints on the digital curve by minimizing simultaneously the number of breakpoints and the approximation error. Using such an approach, the algorithm proposes a set of solutions at its end. The user may choose his own solution according to its objective. The proposed approach is evaluated on curves issued from the literature and compared successfully with many classical approaches. 1.
Asynchronous Master/Slave MOEAs and Heterogeneous Evaluation Costs
, 2012
"... Parallel masterslave evolutionary algorithms easily lead to linear speedups in the case of a small number of nodes...andhomogeneous computational costs of the evaluations. However, modern computer now routinely have several hundreds of nodes – and in many realworld applications in which fitness c ..."
Abstract
 Add to MetaCart
(Show Context)
Parallel masterslave evolutionary algorithms easily lead to linear speedups in the case of a small number of nodes...andhomogeneous computational costs of the evaluations. However, modern computer now routinely have several hundreds of nodes – and in many realworld applications in which fitness computation involves heavy numerical simulations, the computational costs of these simulations can greatly vary from one individual to the next. A simple answer to the latter problem is to use asynchronous steadystate reproduction schemes. But the resulting algorithms then differ from the original sequential version, with two consequences: First, the linear speedup does not hold any more; Second, the convergence might be hindered by the heterogeneity of the evaluation costs. The multiobjective optimization of a diesel engine is first presented, a realworld case study where evaluations require several hours of CPU, and are very heterogeneous in terms of CPU cost. Both the speedup of asynchronous parallel master/slave algorithms in case of large number of nodes, and their convergence toward the Pareto Front in case of heterogeneous computation times, are then experimentally analyzed on artificial test functions. An alternative selection scheme involving the computational cost of the fitness evaluation is then proposed, that counteracts the effects of heterogeneity on convergence toward the Pareto Front.
Author manuscript, published in "ICPR (2), Hong Kong (2006)" Approximation of Digital Curves using a MultiObjective Genetic Algorithm
, 2009
"... In this paper, a digital planar curve approximation method based on a multiobjective genetic algorithm is proposed. In this method, the optimization/exploration algorithm locates breakpoints on the digital curve by minimizing simultaneously the number of breakpoints and the approximation error. Usi ..."
Abstract
 Add to MetaCart
(Show Context)
In this paper, a digital planar curve approximation method based on a multiobjective genetic algorithm is proposed. In this method, the optimization/exploration algorithm locates breakpoints on the digital curve by minimizing simultaneously the number of breakpoints and the approximation error. Using such an approach, the algorithm proposes a set of solutions at its end. The user may choose his own solution according to its objective. The proposed approach is evaluated on curves issued from the literature and compared successfully with many classical approaches. 1.
Polygonal Approximation of Digital Curves Using a MultiObjective Genetic Algorithm
"... In this paper, a polygonal approximation approach based on a multiobjective genetic algorithm is proposed. In this method, the optimization/exploration algorithm locates breakpoints on the digital curve by minimizing simultaneously the number of breakpoints and the approximation error. Using such a ..."
Abstract
 Add to MetaCart
(Show Context)
In this paper, a polygonal approximation approach based on a multiobjective genetic algorithm is proposed. In this method, the optimization/exploration algorithm locates breakpoints on the digital curve by minimizing simultaneously the number of breakpoints and the approximation error. Using such an approach, the algorithm proposes a set of solutions at its end. This set which is called the Pareto Front in the multi objective optimization field contains solutions that represent tradeoffs between the two classical quality criteria of polygonal approximation: the Integral Square Error (ISE) and the number of vertices. The user may choose his own solution according to its objective. The proposed approach is evaluated on curves issued from the literature and compared with many classical approaches. Keywords: Polygonal Approximation, MultiObjective Optimization, Genetic Algorithm, Pareto Front. 1
ALGORITHM: A TOOL FOR FOREST MANAGEMENT
, 2005
"... This paper describes research on the use of a genetic algorithm (GA) to prescribe treatment plans for forest management at the stand level. Forest management refers to making decisions about when and where to intervene in the natural growth of forests to achieve objectives, such as enhancing the vis ..."
Abstract
 Add to MetaCart
This paper describes research on the use of a genetic algorithm (GA) to prescribe treatment plans for forest management at the stand level. Forest management refers to making decisions about when and where to intervene in the natural growth of forests to achieve objectives, such as enhancing the visual quality of a stand or maximizing timber yield. A prescription is a schedule of thinning treatments applied to stands over a planning horizon. When multiple management goals exist treatment prescription becomes a complex multiobjective problem. The effectiveness of a GA depends on selecting an appropriate representation and germane fitness function. This paper discusses design decisions and presents a series of experiments testing the performance of the GA. Different parameter settings are compared and the GA is contrasted with some other heuristic search methods. The final experiment compares a plan created by the GA to a plan recommended by a human expert.
Final Report: SAGA 3 (03DG11244225390)
, 2004
"... This report documents the latest work in the SAGA (Spray Advisor using Genetic Algorithms) project, the goal of which is to provide an effective means for a computer to recommend an optimal set of parameters for spraying a wooded area with pesticide. Current work is in two directions. One of these i ..."
Abstract
 Add to MetaCart
This report documents the latest work in the SAGA (Spray Advisor using Genetic Algorithms) project, the goal of which is to provide an effective means for a computer to recommend an optimal set of parameters for spraying a wooded area with pesticide. Current work is in two directions. One of these is the maintenance of older SAGA programs, some of which were built on nowobsolete platforms. The other is the development of a unified framework, SAGA3, to maximize code reuse and provide a consistent and flexible interface across SAGA programs. Brief History of the SAGA Project The most recent work in the SAGA project builds upon what previous researchers have already accomplished. This section briefly describes the history of the SAGA project and its successes to set the stage for discussion of the issues facing SAGA and how we resolve them. SAGA originated as a prototype generational genetic algorithm (henceforth GA) written in Fortran. To compute the fitness of an individual, it created a text file
Memetic Algorithm Based on a Constraint Satisfaction Technique for VRPTW
"... Abstract. In this paper a Memetic Algorithm (MA) is proposed for ..."
(Show Context)