2006
1.
Artur Chorążyczewski, Roman Galar
Adaptive Saddle Crossings in Diverse Landscapes Proceedings Article
In: Arabas, J. (Ed.): Evolutionary Computation and Global Optimization, pp. 77-85, Warsaw University of Technology Publishing House Warsaw University of Technology Publishing House, 2006, ISSN: 0137-2343.
Abstract | Links | BibTeX | Tagi: kaeiog
@inproceedings{KAEiOG2006.77-85,
title = {Adaptive Saddle Crossings in Diverse Landscapes},
author = {Artur Chorążyczewski and Roman Galar},
editor = {J. Arabas},
url = {https://troja.uksw.edu.pl/pdf/kaeiog/KAEiOG2006.77-85.pdf},
issn = {0137-2343},
year = {2006},
date = {2006-01-01},
booktitle = {Evolutionary Computation and Global Optimization},
number = {156},
pages = {77-85},
publisher = {Warsaw University of Technology Publishing House},
organization = {Warsaw University of Technology Publishing House},
series = {Prace Naukowe, Elektronika},
abstract = {There is always a question as to how sound are assessments of evolutionary algorithms capabilities based on the tests made in the specific adaptive landscapes. We present some results of numerical experiments concerning the ability of phenotypic evolution to cross saddles between adaptive peaks of different shapes. Asexual reproduction, proportional selection and normally distributed mutations are used. Some related comments on the benchmarking of evolutionary algorithms are also given.},
keywords = {kaeiog},
pubstate = {published},
tppubtype = {inproceedings}
}
There is always a question as to how sound are assessments of evolutionary algorithms capabilities based on the tests made in the specific adaptive landscapes. We present some results of numerical experiments concerning the ability of phenotypic evolution to cross saddles between adaptive peaks of different shapes. Asexual reproduction, proportional selection and normally distributed mutations are used. Some related comments on the benchmarking of evolutionary algorithms are also given.
2.
Jarosław Arabas, Andrzej Dydyński
An Algorithm of Incremental Construction of Nonlinear Parametric Approximators Proceedings Article
In: Arabas, J. (Ed.): Evolutionary Computation and Global Optimization, pp. 31-38, Warsaw University of Technology Publishing House Warsaw University of Technology Publishing House, 2006, ISSN: 0137-2343.
Abstract | Links | BibTeX | Tagi: kaeiog
@inproceedings{KAEiOG2006.31-38,
title = {An Algorithm of Incremental Construction of Nonlinear Parametric Approximators},
author = {Jarosław Arabas and Andrzej Dydyński},
editor = {J. Arabas},
url = {https://troja.uksw.edu.pl/pdf/kaeiog/KAEiOG2006.31-38.pdf},
issn = {0137-2343},
year = {2006},
date = {2006-01-01},
booktitle = {Evolutionary Computation and Global Optimization},
number = {156},
pages = {31-38},
publisher = {Warsaw University of Technology Publishing House},
organization = {Warsaw University of Technology Publishing House},
series = {Prace Naukowe, Elektronika},
abstract = {A novel algorithm of incremental construction of nonlinear parametric approximators is introduced that needs a relatively small number of parameters to achieve good accuracy. The approximator is defined as a linear combination of nonlinear base functions, and the method adds the base functions one by one, observing the correlation between the residue function and the candidate for the base function. The base functions do not need to be homogenous, i.e. to share the same formula. The method was experimentally verified and compared to neural networks, and the preliminary results were encouraging enough to develop the presented approach further.},
keywords = {kaeiog},
pubstate = {published},
tppubtype = {inproceedings}
}
A novel algorithm of incremental construction of nonlinear parametric approximators is introduced that needs a relatively small number of parameters to achieve good accuracy. The approximator is defined as a linear combination of nonlinear base functions, and the method adds the base functions one by one, observing the correlation between the residue function and the candidate for the base function. The base functions do not need to be homogenous, i.e. to share the same formula. The method was experimentally verified and compared to neural networks, and the preliminary results were encouraging enough to develop the presented approach further.
3.
Marek Kubiak
Analysis of distance between vehicle routing problem solutions generated by memetic algorithms Proceedings Article
In: Arabas, J. (Ed.): Evolutionary Computation and Global Optimization, pp. 223-236, Warsaw University of Technology Publishing House Warsaw University of Technology Publishing House, 2006, ISSN: 0137-2343.
Abstract | Links | BibTeX | Tagi: kaeiog
@inproceedings{KAEiOG2006.223-236,
title = {Analysis of distance between vehicle routing problem solutions generated by memetic algorithms},
author = {Marek Kubiak},
editor = {J. Arabas},
url = {https://troja.uksw.edu.pl/pdf/kaeiog/KAEiOG2006.223-236.pdf},
issn = {0137-2343},
year = {2006},
date = {2006-01-01},
booktitle = {Evolutionary Computation and Global Optimization},
number = {156},
pages = {223-236},
publisher = {Warsaw University of Technology Publishing House},
organization = {Warsaw University of Technology Publishing House},
series = {Prace Naukowe, Elektronika},
abstract = {This paper focuses on analysis of distance between solutions of the capacitated vehicle routing problem generated by memetic algorihms with different crossover operators. The goal of the analysis is to see what are relative positions of such solutions in search spaces and if different algorithms explore different parts of these spaces. In the described memetic algorithms five different crossover operators are used. The conducted computational experiment shows that solutions generated by the algorithms have very similar, very good quality. From the distance analysis it appears that there are two types of instances of the considered problem: 10 instances have very similar solutions, concentrated in small regions of spaces (textquoterightbig valleystextquoteright), while other 12 have good solutions which reside in different parts of search spaces, implying wide and flat regions of good solutions.},
keywords = {kaeiog},
pubstate = {published},
tppubtype = {inproceedings}
}
This paper focuses on analysis of distance between solutions of the capacitated vehicle routing problem generated by memetic algorihms with different crossover operators. The goal of the analysis is to see what are relative positions of such solutions in search spaces and if different algorithms explore different parts of these spaces. In the described memetic algorithms five different crossover operators are used. The conducted computational experiment shows that solutions generated by the algorithms have very similar, very good quality. From the distance analysis it appears that there are two types of instances of the considered problem: 10 instances have very similar solutions, concentrated in small regions of spaces (textquoterightbig valleystextquoteright), while other 12 have good solutions which reside in different parts of search spaces, implying wide and flat regions of good solutions.
4.
Joanna Kołodziej
An Application of Erlang Mixture Distributions to modeling the reproduction mechanism in estimation of distribution algorithms. Proceedings Article
In: Arabas, J. (Ed.): Evolutionary Computation and Global Optimization, pp. 203-210, Warsaw University of Technology Publishing House Warsaw University of Technology Publishing House, 2006, ISSN: 0137-2343.
Abstract | Links | BibTeX | Tagi: kaeiog
@inproceedings{KAEiOG2006.203-210,
title = {An Application of Erlang Mixture Distributions to modeling the reproduction mechanism in estimation of distribution algorithms.},
author = {Joanna Kołodziej},
editor = {J. Arabas},
url = {https://troja.uksw.edu.pl/pdf/kaeiog/KAEiOG2006.203-210.pdf},
issn = {0137-2343},
year = {2006},
date = {2006-01-01},
booktitle = {Evolutionary Computation and Global Optimization},
number = {156},
pages = {203-210},
publisher = {Warsaw University of Technology Publishing House},
organization = {Warsaw University of Technology Publishing House},
series = {Prace Naukowe, Elektronika},
abstract = {We present the idea of an application of the mixtures of Erlang distributions in the construction of the recombination mechanism in estimation of distribution algorithms. We analyze main properties of Erlang mixtures and define a new Erlang Mixture Estimation of Distribution Algorithm. We try to compare the efficiencies of ErM-EDA and evolutionary strategy in case of large populations. Some experimental results are presented after simple theoretical studies.},
keywords = {kaeiog},
pubstate = {published},
tppubtype = {inproceedings}
}
We present the idea of an application of the mixtures of Erlang distributions in the construction of the recombination mechanism in estimation of distribution algorithms. We analyze main properties of Erlang mixtures and define a new Erlang Mixture Estimation of Distribution Algorithm. We try to compare the efficiencies of ErM-EDA and evolutionary strategy in case of large populations. Some experimental results are presented after simple theoretical studies.
5.
Jarosław Arabas
Asymptotic Correctness of the Evolutionary Algorithm in Banach Space Proceedings Article
In: Arabas, J. (Ed.): Evolutionary Computation and Global Optimization, pp. 21-30, Warsaw University of Technology Publishing House Warsaw University of Technology Publishing House, 2006, ISSN: 0137-2343.
Abstract | Links | BibTeX | Tagi: kaeiog
@inproceedings{KAEiOG2006.21-30,
title = {Asymptotic Correctness of the Evolutionary Algorithm in Banach Space},
author = {Jarosław Arabas},
editor = {J. Arabas},
url = {https://troja.uksw.edu.pl/pdf/kaeiog/KAEiOG2006.21-30.pdf},
issn = {0137-2343},
year = {2006},
date = {2006-01-01},
booktitle = {Evolutionary Computation and Global Optimization},
number = {156},
pages = {21-30},
publisher = {Warsaw University of Technology Publishing House},
organization = {Warsaw University of Technology Publishing House},
series = {Prace Naukowe, Elektronika},
abstract = {The paper gives two theorems for the asymptotic correctness of an evolutionary algorithm (EA) that processes chromosomes from an arbitrary Banach space. It is argued that even if the mutation cannot yield an arbitrarily far offspring, the EA may be asymptotically correct provided that the selection is nonelitist and each feasible individual may reproduce with nonzero probability. An illustrative example accompanies the paper.},
keywords = {kaeiog},
pubstate = {published},
tppubtype = {inproceedings}
}
The paper gives two theorems for the asymptotic correctness of an evolutionary algorithm (EA) that processes chromosomes from an arbitrary Banach space. It is argued that even if the mutation cannot yield an arbitrarily far offspring, the EA may be asymptotically correct provided that the selection is nonelitist and each feasible individual may reproduce with nonzero probability. An illustrative example accompanies the paper.
6.
Eugene Eberbach
Capturing Evolution of Polymorphic Viruses Proceedings Article
In: Arabas, J. (Ed.): Evolutionary Computation and Global Optimization, pp. 125-138, Warsaw University of Technology Publishing House Warsaw University of Technology Publishing House, 2006, ISSN: 0137-2343.
Abstract | Links | BibTeX | Tagi: kaeiog
@inproceedings{KAEiOG2006.125-138,
title = {Capturing Evolution of Polymorphic Viruses},
author = {Eugene Eberbach},
editor = {J. Arabas},
url = {https://troja.uksw.edu.pl/pdf/kaeiog/KAEiOG2006.125-138.pdf},
issn = {0137-2343},
year = {2006},
date = {2006-01-01},
booktitle = {Evolutionary Computation and Global Optimization},
number = {156},
pages = {125-138},
publisher = {Warsaw University of Technology Publishing House},
organization = {Warsaw University of Technology Publishing House},
series = {Prace Naukowe, Elektronika},
abstract = {In this paper three models of polymorphic viruses are presented. These models had to capture self-poduction, evolution and damaging payload of polymorphic viruses. They have been derived using a formal model of Evolutionary Computation the Evolutionary Turing Machine, cellular space models, and the $-calculus process algebra for problem solving. Some preliminary results associated with these models are discussed.},
keywords = {kaeiog},
pubstate = {published},
tppubtype = {inproceedings}
}
In this paper three models of polymorphic viruses are presented. These models had to capture self-poduction, evolution and damaging payload of polymorphic viruses. They have been derived using a formal model of Evolutionary Computation the Evolutionary Turing Machine, cellular space models, and the $-calculus process algebra for problem solving. Some preliminary results associated with these models are discussed.
7.
Krzysztof Trojanowski
Clonal Selection Principle Based Approach to Non-Stationary Optimization Tasks Proceedings Article
In: Arabas, J. (Ed.): Evolutionary Computation and Global Optimization, pp. 375-384, Warsaw University of Technology Publishing House Warsaw University of Technology Publishing House, 2006, ISSN: 0137-2343.
@inproceedings{KAEiOG2006.375-384,
title = {Clonal Selection Principle Based Approach to Non-Stationary Optimization Tasks},
author = {Krzysztof Trojanowski},
editor = {J. Arabas},
url = {https://troja.uksw.edu.pl/pdf/kaeiog/KAEiOG2006.375-384.pdf},
issn = {0137-2343},
year = {2006},
date = {2006-01-01},
booktitle = {Evolutionary Computation and Global Optimization},
number = {156},
pages = {375-384},
publisher = {Warsaw University of Technology Publishing House},
organization = {Warsaw University of Technology Publishing House},
series = {Prace Naukowe, Elektronika},
keywords = {kaeiog},
pubstate = {published},
tppubtype = {inproceedings}
}
8.
Urszula Boryczka, Rafał Skinderowicz, Damian Świstowski
Comparative Study: ACO and EC for TSP Proceedings Article
In: Arabas, J. (Ed.): Evolutionary Computation and Global Optimization, pp. 69-76, Warsaw University of Technology Publishing House Warsaw University of Technology Publishing House, 2006, ISSN: 0137-2343.
Abstract | Links | BibTeX | Tagi: kaeiog
@inproceedings{KAEiOG2006.69-76,
title = {Comparative Study: ACO and EC for TSP},
author = {Urszula Boryczka and Rafał Skinderowicz and Damian Świstowski},
editor = {J. Arabas},
url = {https://troja.uksw.edu.pl/pdf/kaeiog/KAEiOG2006.69-76.pdf},
issn = {0137-2343},
year = {2006},
date = {2006-01-01},
booktitle = {Evolutionary Computation and Global Optimization},
number = {156},
pages = {69-76},
publisher = {Warsaw University of Technology Publishing House},
organization = {Warsaw University of Technology Publishing House},
series = {Prace Naukowe, Elektronika},
abstract = {Evolutionary Computing (EC) and Ant Colony Optimization (ACO) apply stochastic searching, parallel investigation as well as autocatalitic process (or stigmergy) to solve optimization problems. This paper concentrates on the Traveling Salesman Problem (TSP) solved by evolutionary and ACO algorithms. We consider the sets of parameters and operators which influence the acting of these algorithms. Two algorithmic structures emphasizing the selection problem are discussed. We describe experiments performed for different instances of TSP problems. The comparison concludes that evolution, which is exploited especially in evolutionary algorithms, can also be observed in the performance of the ACO approach.},
keywords = {kaeiog},
pubstate = {published},
tppubtype = {inproceedings}
}
Evolutionary Computing (EC) and Ant Colony Optimization (ACO) apply stochastic searching, parallel investigation as well as autocatalitic process (or stigmergy) to solve optimization problems. This paper concentrates on the Traveling Salesman Problem (TSP) solved by evolutionary and ACO algorithms. We consider the sets of parameters and operators which influence the acting of these algorithms. Two algorithmic structures emphasizing the selection problem are discussed. We describe experiments performed for different instances of TSP problems. The comparison concludes that evolution, which is exploited especially in evolutionary algorithms, can also be observed in the performance of the ACO approach.
9.
Piotr Kaczyński, Paweł D. Domański
Controller adaptation using dynamic evolutionary algorithm Proceedings Article
In: Arabas, J. (Ed.): Evolutionary Computation and Global Optimization, pp. 179-187, Warsaw University of Technology Publishing House Warsaw University of Technology Publishing House, 2006, ISSN: 0137-2343.
Abstract | Links | BibTeX | Tagi: kaeiog
@inproceedings{KAEiOG2006.179-187,
title = {Controller adaptation using dynamic evolutionary algorithm},
author = {Piotr Kaczyński and Paweł D. Domański},
editor = {J. Arabas},
url = {https://troja.uksw.edu.pl/pdf/kaeiog/KAEiOG2006.179-187.pdf},
issn = {0137-2343},
year = {2006},
date = {2006-01-01},
booktitle = {Evolutionary Computation and Global Optimization},
number = {156},
pages = {179-187},
publisher = {Warsaw University of Technology Publishing House},
organization = {Warsaw University of Technology Publishing House},
series = {Prace Naukowe, Elektronika},
abstract = {In this paper we present an algorithmic solution to PID controller parameters tuning task. Firstly, the brief test model and a simple example on how to tune PID parameters using genetic algorithm are presented. This functionality is then tested in more complicated situation of a non-stationary linear object. Adaptivity of such a solution is considered along with some results.},
keywords = {kaeiog},
pubstate = {published},
tppubtype = {inproceedings}
}
In this paper we present an algorithmic solution to PID controller parameters tuning task. Firstly, the brief test model and a simple example on how to tune PID parameters using genetic algorithm are presented. This functionality is then tested in more complicated situation of a non-stationary linear object. Adaptivity of such a solution is considered along with some results.
10.
Przemysław Dziemieszkiewicz
Distribution of Populations Generated by Multi-Objective Optimization Evolutionary Algorithms Proceedings Article
In: Arabas, J. (Ed.): Evolutionary Computation and Global Optimization, pp. 117-124, Warsaw University of Technology Publishing House Warsaw University of Technology Publishing House, 2006, ISSN: 0137-2343.
Abstract | Links | BibTeX | Tagi: kaeiog
@inproceedings{KAEiOG2006.117-124,
title = {Distribution of Populations Generated by Multi-Objective Optimization Evolutionary Algorithms},
author = {Przemysław Dziemieszkiewicz},
editor = {J. Arabas},
url = {https://troja.uksw.edu.pl/pdf/kaeiog/KAEiOG2006.117-124.pdf},
issn = {0137-2343},
year = {2006},
date = {2006-01-01},
booktitle = {Evolutionary Computation and Global Optimization},
number = {156},
pages = {117-124},
publisher = {Warsaw University of Technology Publishing House},
organization = {Warsaw University of Technology Publishing House},
series = {Prace Naukowe, Elektronika},
abstract = {This paper considers the distribution of populations generated by multi-objective evolutionary algorithms. Some algorithms have a tendency to produce multimodal distributions of population while other algorithms produce a compact cluster of solutions. Particular MOEAs have a tendency to one of these distributions. This paper will show that one feature of an algorithm defines it – the selection method. If selection is based on the evaluation of parents, the algorithm generates a compact population every time. When the selection is based on the fitness of the offspring, the algorithm is able to generate bimodal distribution of population. Five MOEA algorithms were analyzed: VEGA, two types of tournament and two methods proposed by the author (MOEA with competitive selection and MOEA with protective selection). The results of the simulation confirmed that, depending on the selection method selected, the analyzed algorithms generate the expected distribution of population. This paper does not assess which algorithm is better or worse but merely seeks to explain the reason for the algorithmstextquoteright properties.},
keywords = {kaeiog},
pubstate = {published},
tppubtype = {inproceedings}
}
This paper considers the distribution of populations generated by multi-objective evolutionary algorithms. Some algorithms have a tendency to produce multimodal distributions of population while other algorithms produce a compact cluster of solutions. Particular MOEAs have a tendency to one of these distributions. This paper will show that one feature of an algorithm defines it – the selection method. If selection is based on the evaluation of parents, the algorithm generates a compact population every time. When the selection is based on the fitness of the offspring, the algorithm is able to generate bimodal distribution of population. Five MOEA algorithms were analyzed: VEGA, two types of tournament and two methods proposed by the author (MOEA with competitive selection and MOEA with protective selection). The results of the simulation confirmed that, depending on the selection method selected, the analyzed algorithms generate the expected distribution of population. This paper does not assess which algorithm is better or worse but merely seeks to explain the reason for the algorithmstextquoteright properties.
11.
Tomasz Gwizdałła
The Efficiency of Evolutionary Optimization of Magnetic Properties of Ternary Alloy Proceedings Article
In: Arabas, J. (Ed.): Evolutionary Computation and Global Optimization, pp. 155-162, Warsaw University of Technology Publishing House Warsaw University of Technology Publishing House, 2006, ISSN: 0137-2343.
Abstract | Links | BibTeX | Tagi: kaeiog
@inproceedings{KAEiOG2006.155-162,
title = {The Efficiency of Evolutionary Optimization of Magnetic Properties of Ternary Alloy},
author = {Tomasz Gwizdałła},
editor = {J. Arabas},
url = {https://troja.uksw.edu.pl/pdf/kaeiog/KAEiOG2006.155-162.pdf},
issn = {0137-2343},
year = {2006},
date = {2006-01-01},
booktitle = {Evolutionary Computation and Global Optimization},
number = {156},
pages = {155-162},
publisher = {Warsaw University of Technology Publishing House},
organization = {Warsaw University of Technology Publishing House},
series = {Prace Naukowe, Elektronika},
abstract = {The area of magnetism offers a wealth of possibilities to study differ-ent algorithmic approaches. The reason is the diversity of different forms which can be used to describe magnetic samples. One can start from the most simple Ising-like system which, being of physical origin, became an inspiration for many models in a various areas of science and the formulation of which fits ideally the earliest representations of genetic algorithms. The recent problems concern elab-orate molecular structures, like eg. spin glasses where the character of interaction between the spins of the same type may be different depending on some random variable or mixed alloys which can have different stoichiometry or structure and therefore present visibly distinct magnetic characteristics. In this paper we will deal with the mixed magnetic ternary alloy - an analog of prussian blue. Some results concerning the physical conclusions coming from applying the evolutionary computation to analyse this material have already been published and here the attention will be paid to the influence of parameter change on the efficiency of procedure.},
keywords = {kaeiog},
pubstate = {published},
tppubtype = {inproceedings}
}
The area of magnetism offers a wealth of possibilities to study differ-ent algorithmic approaches. The reason is the diversity of different forms which can be used to describe magnetic samples. One can start from the most simple Ising-like system which, being of physical origin, became an inspiration for many models in a various areas of science and the formulation of which fits ideally the earliest representations of genetic algorithms. The recent problems concern elab-orate molecular structures, like eg. spin glasses where the character of interaction between the spins of the same type may be different depending on some random variable or mixed alloys which can have different stoichiometry or structure and therefore present visibly distinct magnetic characteristics. In this paper we will deal with the mixed magnetic ternary alloy - an analog of prussian blue. Some results concerning the physical conclusions coming from applying the evolutionary computation to analyse this material have already been published and here the attention will be paid to the influence of parameter change on the efficiency of procedure.
12.
Piotr Orantek, Tadeusz Burczyński
The evolutionary algorithm in stochastic optimization and identification problems Proceedings Article
In: Arabas, J. (Ed.): Evolutionary Computation and Global Optimization, pp. 309-319, Warsaw University of Technology Publishing House Warsaw University of Technology Publishing House, 2006, ISSN: 0137-2343.
Abstract | Links | BibTeX | Tagi: kaeiog
@inproceedings{KAEiOG2006.309-319,
title = {The evolutionary algorithm in stochastic optimization and identification problems},
author = {Piotr Orantek and Tadeusz Burczyński},
editor = {J. Arabas},
url = {https://troja.uksw.edu.pl/pdf/kaeiog/KAEiOG2006.309-319.pdf},
issn = {0137-2343},
year = {2006},
date = {2006-01-01},
booktitle = {Evolutionary Computation and Global Optimization},
number = {156},
pages = {309-319},
publisher = {Warsaw University of Technology Publishing House},
organization = {Warsaw University of Technology Publishing House},
series = {Prace Naukowe, Elektronika},
abstract = {This paper is devoted to the application of evolutionary computing in optimization and identification problems in uncertain random conditions. The algorithm is based on the stochastic representation of the data. Chromosomes are represented by multidimensional random vectors consisting of random genes in the form of independent random variables with the Gaussian density probability function. The stochastic optimization problem is replaced by a deterministic one by evolutionary computing for vector genes consisting of mean values and standard deviations. Special operators for mutation, crososver and selection are proposed. Two numerical tests are presented.},
keywords = {kaeiog},
pubstate = {published},
tppubtype = {inproceedings}
}
This paper is devoted to the application of evolutionary computing in optimization and identification problems in uncertain random conditions. The algorithm is based on the stochastic representation of the data. Chromosomes are represented by multidimensional random vectors consisting of random genes in the form of independent random variables with the Gaussian density probability function. The stochastic optimization problem is replaced by a deterministic one by evolutionary computing for vector genes consisting of mean values and standard deviations. Special operators for mutation, crososver and selection are proposed. Two numerical tests are presented.
13.
Henryk Potrzebowski, Jarosław Stańczak, Krzysztof Sęp
Evolutionary Algorithm to Find Graph Covering Subsets Using a-Cliques Proceedings Article
In: Arabas, J. (Ed.): Evolutionary Computation and Global Optimization, pp. 351-358, Warsaw University of Technology Publishing House Warsaw University of Technology Publishing House, 2006, ISSN: 0137-2343.
Abstract | Links | BibTeX | Tagi: kaeiog
@inproceedings{KAEiOG2006.351-358,
title = {Evolutionary Algorithm to Find Graph Covering Subsets Using a-Cliques},
author = {Henryk Potrzebowski and Jarosław Stańczak and Krzysztof Sęp},
editor = {J. Arabas},
url = {https://troja.uksw.edu.pl/pdf/kaeiog/KAEiOG2006.351-358.pdf},
issn = {0137-2343},
year = {2006},
date = {2006-01-01},
booktitle = {Evolutionary Computation and Global Optimization},
number = {156},
pages = {351-358},
publisher = {Warsaw University of Technology Publishing House},
organization = {Warsaw University of Technology Publishing House},
series = {Prace Naukowe, Elektronika},
abstract = {The article describes a new evolutionary based method to divide graph into strongly connected structures we called a-cliques. The a-clique is a generalization of a clique concept with the introduction of parameter a. Using this parameter it is possible to control the degree (or strength) of connections among vertices (nodes) of this sub-graph structure. The evolutionary approach is proposed as a method that enables to find separate a-cliques that cover the set of graph vertices.},
keywords = {kaeiog},
pubstate = {published},
tppubtype = {inproceedings}
}
The article describes a new evolutionary based method to divide graph into strongly connected structures we called a-cliques. The a-clique is a generalization of a clique concept with the introduction of parameter a. Using this parameter it is possible to control the degree (or strength) of connections among vertices (nodes) of this sub-graph structure. The evolutionary approach is proposed as a method that enables to find separate a-cliques that cover the set of graph vertices.
14.
Rafał Biedrzycki, Jarosław Arabas
Evolutionary and greedy exploration of the space of decision trees Proceedings Article
In: Arabas, J. (Ed.): Evolutionary Computation and Global Optimization, pp. 61-68, Warsaw University of Technology Publishing House Warsaw University of Technology Publishing House, 2006, ISSN: 0137-2343.
Abstract | Links | BibTeX | Tagi: kaeiog
@inproceedings{KAEiOG2006.61-68,
title = {Evolutionary and greedy exploration of the space of decision trees},
author = {Rafał Biedrzycki and Jarosław Arabas},
editor = {J. Arabas},
url = {https://troja.uksw.edu.pl/pdf/kaeiog/KAEiOG2006.61-68.pdf},
issn = {0137-2343},
year = {2006},
date = {2006-01-01},
booktitle = {Evolutionary Computation and Global Optimization},
number = {156},
pages = {61-68},
publisher = {Warsaw University of Technology Publishing House},
organization = {Warsaw University of Technology Publishing House},
series = {Prace Naukowe, Elektronika},
abstract = {This paper addresses the issue of the decision trees induction. We define the space of all possible trees and try to find good trees by searching that space. We compare performance of an evolutionary algorithm and standard, problem-specific algorithms (ID3, C4.5).},
keywords = {kaeiog},
pubstate = {published},
tppubtype = {inproceedings}
}
This paper addresses the issue of the decision trees induction. We define the space of all possible trees and try to find good trees by searching that space. We compare performance of an evolutionary algorithm and standard, problem-specific algorithms (ID3, C4.5).
15.
Anna Pagacz, Gunther Raidl, Stanisław Zawiślak
Evolutionary Approach to Constrained Minimum Spanning Tree Problem - Commercial Software Based Application Proceedings Article
In: Arabas, J. (Ed.): Evolutionary Computation and Global Optimization, pp. 331-341, Warsaw University of Technology Publishing House Warsaw University of Technology Publishing House, 2006, ISSN: 0137-2343.
Abstract | Links | BibTeX | Tagi: kaeiog
@inproceedings{KAEiOG2006.331-341,
title = {Evolutionary Approach to Constrained Minimum Spanning Tree Problem - Commercial Software Based Application},
author = {Anna Pagacz and Gunther Raidl and Stanisław Zawiślak},
editor = {J. Arabas},
url = {https://troja.uksw.edu.pl/pdf/kaeiog/KAEiOG2006.331-341.pdf},
issn = {0137-2343},
year = {2006},
date = {2006-01-01},
booktitle = {Evolutionary Computation and Global Optimization},
number = {156},
pages = {331-341},
publisher = {Warsaw University of Technology Publishing House},
organization = {Warsaw University of Technology Publishing House},
series = {Prace Naukowe, Elektronika},
abstract = {The constrained minimum spanning tree problem is considered in the paper. We assume that the degree of any vertex should not exceed a particular constraint d. In this formulation the problem turns into NP-hard one, therefore the evolutionary approach is applicable. The edge set representation of a chromosome was utilized for a tree in the algorithm. The evolutionary algorithm was worked out and the related computer program has been written. Interfaces between the core program and MS Visio as well as the data base system were prepared. The results obtained by means of the system are shown.},
keywords = {kaeiog},
pubstate = {published},
tppubtype = {inproceedings}
}
The constrained minimum spanning tree problem is considered in the paper. We assume that the degree of any vertex should not exceed a particular constraint d. In this formulation the problem turns into NP-hard one, therefore the evolutionary approach is applicable. The edge set representation of a chromosome was utilized for a tree in the algorithm. The evolutionary algorithm was worked out and the related computer program has been written. Interfaces between the core program and MS Visio as well as the data base system were prepared. The results obtained by means of the system are shown.
16.
Przemysław Prętki, Andrzej Obuchowicz
Evolutionary Bands for the Expected Response in Nonlinear Regression Proceedings Article
In: Arabas, J. (Ed.): Evolutionary Computation and Global Optimization, pp. 359-364, Warsaw University of Technology Publishing House Warsaw University of Technology Publishing House, 2006, ISSN: 0137-2343.
Abstract | Links | BibTeX | Tagi: kaeiog
@inproceedings{KAEiOG2006.359-364,
title = {Evolutionary Bands for the Expected Response in Nonlinear Regression},
author = {Przemysław Prętki and Andrzej Obuchowicz},
editor = {J. Arabas},
url = {https://troja.uksw.edu.pl/pdf/kaeiog/KAEiOG2006.359-364.pdf},
issn = {0137-2343},
year = {2006},
date = {2006-01-01},
booktitle = {Evolutionary Computation and Global Optimization},
number = {156},
pages = {359-364},
publisher = {Warsaw University of Technology Publishing House},
organization = {Warsaw University of Technology Publishing House},
series = {Prace Naukowe, Elektronika},
abstract = {In this paper evolutionary algorithms are applied to computation of confidence intervals for the expected response of nonlinear models. A simple phe-notypic evolutionary algorithm was adapted to deal with nonlinear constraints and utilized to find the maximum and minimum value of a nonlinear model responses inside a confidence region. Moreover, the adequacy of the proposed approach is tested in a series of numerical simulations, and compared with the commonly applied linearization technique.},
keywords = {kaeiog},
pubstate = {published},
tppubtype = {inproceedings}
}
In this paper evolutionary algorithms are applied to computation of confidence intervals for the expected response of nonlinear models. A simple phe-notypic evolutionary algorithm was adapted to deal with nonlinear constraints and utilized to find the maximum and minimum value of a nonlinear model responses inside a confidence region. Moreover, the adequacy of the proposed approach is tested in a series of numerical simulations, and compared with the commonly applied linearization technique.
17.
Anna Jasińska-Suwada
The Evolutionary Heuristics Applied to the Analysis of Multidimensional Data Proceedings Article
In: Arabas, J. (Ed.): Evolutionary Computation and Global Optimization, pp. 163-170, Warsaw University of Technology Publishing House Warsaw University of Technology Publishing House, 2006, ISSN: 0137-2343.
Abstract | Links | BibTeX | Tagi: kaeiog
@inproceedings{KAEiOG2006.163-170,
title = {The Evolutionary Heuristics Applied to the Analysis of Multidimensional Data},
author = {Anna Jasińska-Suwada},
editor = {J. Arabas},
url = {https://troja.uksw.edu.pl/pdf/kaeiog/KAEiOG2006.163-170.pdf},
issn = {0137-2343},
year = {2006},
date = {2006-01-01},
booktitle = {Evolutionary Computation and Global Optimization},
number = {156},
pages = {163-170},
publisher = {Warsaw University of Technology Publishing House},
organization = {Warsaw University of Technology Publishing House},
series = {Prace Naukowe, Elektronika},
abstract = {This paper describes the pattern recognition system for analysis of multidimensional data, based on natural meta-heuristics. The system consists of tree modules: preprocessing, feature extraction and clustering. Feature extraction module is based on Molecular Dynamic (MD). In clustering are used two natural methods: Simulated Annealing (SA) and Taboo Search (TS). The system is used to analyze an evolving population of individuals equipped with textquoterightgenetic codestextquoteright. Clustering module extracts groups of data with similar genetic code named clusters and make of possible to observe their geographical localization. The feature extraction verifies the clustering and allows analyzing of clustering patterns, their shapes and the distances between them.},
keywords = {kaeiog},
pubstate = {published},
tppubtype = {inproceedings}
}
This paper describes the pattern recognition system for analysis of multidimensional data, based on natural meta-heuristics. The system consists of tree modules: preprocessing, feature extraction and clustering. Feature extraction module is based on Molecular Dynamic (MD). In clustering are used two natural methods: Simulated Annealing (SA) and Taboo Search (TS). The system is used to analyze an evolving population of individuals equipped with textquoterightgenetic codestextquoteright. Clustering module extracts groups of data with similar genetic code named clusters and make of possible to observe their geographical localization. The feature extraction verifies the clustering and allows analyzing of clustering patterns, their shapes and the distances between them.
18.
Witold Beluch, Tadeusz Burczyński, Wacław Kuś
Evolutionary Optimization and Identification of Hybrid Laminates Proceedings Article
In: Arabas, J. (Ed.): Evolutionary Computation and Global Optimization, pp. 39-47, Warsaw University of Technology Publishing House Warsaw University of Technology Publishing House, 2006, ISSN: 0137-2343.
Abstract | Links | BibTeX | Tagi: kaeiog
@inproceedings{KAEiOG2006.39-47,
title = {Evolutionary Optimization and Identification of Hybrid Laminates},
author = {Witold Beluch and Tadeusz Burczyński and Wacław Kuś},
editor = {J. Arabas},
url = {https://troja.uksw.edu.pl/pdf/kaeiog/KAEiOG2006.39-47.pdf},
issn = {0137-2343},
year = {2006},
date = {2006-01-01},
booktitle = {Evolutionary Computation and Global Optimization},
number = {156},
pages = {39-47},
publisher = {Warsaw University of Technology Publishing House},
organization = {Warsaw University of Technology Publishing House},
series = {Prace Naukowe, Elektronika},
abstract = {The paper deals with the optimization and identification of hybrid fibre-reinforced laminates. Two tasks are considered: i) the ply orientations optimization, ii) the material constants identification. Different optimization and identification criteria connected with the modal analysis are considered. The evolutionary algorithm is used as the optimization and identification method. In order to accelerate the calculations, the distributed version of the evolutionary algorithm is employed. Finite element method software is used to solve the direct problem for the laminates. Numerical examples of optimization and identification are attached.},
keywords = {kaeiog},
pubstate = {published},
tppubtype = {inproceedings}
}
The paper deals with the optimization and identification of hybrid fibre-reinforced laminates. Two tasks are considered: i) the ply orientations optimization, ii) the material constants identification. Different optimization and identification criteria connected with the modal analysis are considered. The evolutionary algorithm is used as the optimization and identification method. In order to accelerate the calculations, the distributed version of the evolutionary algorithm is employed. Finite element method software is used to solve the direct problem for the laminates. Numerical examples of optimization and identification are attached.
19.
Wacław Kuś, Tadeusz Burczyński
Evolutionary Optimization using Enterprise Grid Based on Alchemi Framework Proceedings Article
In: Arabas, J. (Ed.): Evolutionary Computation and Global Optimization, pp. 257-262, Warsaw University of Technology Publishing House Warsaw University of Technology Publishing House, 2006, ISSN: 0137-2343.
Abstract | Links | BibTeX | Tagi: kaeiog
@inproceedings{KAEiOG2006.257-262,
title = {Evolutionary Optimization using Enterprise Grid Based on Alchemi Framework},
author = {Wacław Kuś and Tadeusz Burczyński},
editor = {J. Arabas},
url = {https://troja.uksw.edu.pl/pdf/kaeiog/KAEiOG2006.257-262.pdf},
issn = {0137-2343},
year = {2006},
date = {2006-01-01},
booktitle = {Evolutionary Computation and Global Optimization},
number = {156},
pages = {257-262},
publisher = {Warsaw University of Technology Publishing House},
organization = {Warsaw University of Technology Publishing House},
series = {Prace Naukowe, Elektronika},
abstract = {This paper describes the evolutionary optimization performed using enterprise grid approach. The Alchemi framework is used. The Alchemi is built on top of .NET framework. The evolutionary algorithm works in parallel form and uses Alchemi Executors to perform computations of fitness function values for chromosomes. The experimental tests measuring Alchemi overhead during computations are presented.},
keywords = {kaeiog},
pubstate = {published},
tppubtype = {inproceedings}
}
This paper describes the evolutionary optimization performed using enterprise grid approach. The Alchemi framework is used. The Alchemi is built on top of .NET framework. The evolutionary algorithm works in parallel form and uses Alchemi Executors to perform computations of fitness function values for chromosomes. The experimental tests measuring Alchemi overhead during computations are presented.
20.
Marcin Witczak, Józef Korbicz
Fault diagnosis of dynamical systems using evolutionary algorithms Proceedings Article
In: Arabas, J. (Ed.): Evolutionary Computation and Global Optimization, pp. 413-420, Warsaw University of Technology Publishing House Warsaw University of Technology Publishing House, 2006, ISSN: 0137-2343.
Abstract | Links | BibTeX | Tagi: kaeiog
@inproceedings{KAEiOG2006.413-420,
title = {Fault diagnosis of dynamical systems using evolutionary algorithms},
author = {Marcin Witczak and Józef Korbicz},
editor = {J. Arabas},
url = {https://troja.uksw.edu.pl/pdf/kaeiog/KAEiOG2006.413-420.pdf},
issn = {0137-2343},
year = {2006},
date = {2006-01-01},
booktitle = {Evolutionary Computation and Global Optimization},
number = {156},
pages = {413-420},
publisher = {Warsaw University of Technology Publishing House},
organization = {Warsaw University of Technology Publishing House},
series = {Prace Naukowe, Elektronika},
abstract = {Challenging and complex design problems arise regularly in modern fault diagnosis systems. Unfortunately, the classical analytical techniques cannot often provide acceptable solutions to such difficult tasks. This explains why soft computing techniques such as evolutionary algorithms are becoming more and more popular in industrial applications of fault diagnosis. The main objective of this paper is to present recent developments regarding the application of evolu-tionary algorithms in fault diagnosis. The main attention is on the techniques that integrate the classical and evolutionary approaches. A selected example, dealing with the DAMADICS benchmark, is carefully described in the paper.},
keywords = {kaeiog},
pubstate = {published},
tppubtype = {inproceedings}
}
Challenging and complex design problems arise regularly in modern fault diagnosis systems. Unfortunately, the classical analytical techniques cannot often provide acceptable solutions to such difficult tasks. This explains why soft computing techniques such as evolutionary algorithms are becoming more and more popular in industrial applications of fault diagnosis. The main objective of this paper is to present recent developments regarding the application of evolu-tionary algorithms in fault diagnosis. The main attention is on the techniques that integrate the classical and evolutionary approaches. A selected example, dealing with the DAMADICS benchmark, is carefully described in the paper.
21.
Wojciech Szynkiewicz
A Game-Theoretic Model of Multi-Robot Interactions Proceedings Article
In: Arabas, J. (Ed.): Evolutionary Computation and Global Optimization, pp. 365-374, Warsaw University of Technology Publishing House Warsaw University of Technology Publishing House, 2006, ISSN: 0137-2343.
Abstract | Links | BibTeX | Tagi: kaeiog
@inproceedings{KAEiOG2006.365-374,
title = {A Game-Theoretic Model of Multi-Robot Interactions},
author = {Wojciech Szynkiewicz},
editor = {J. Arabas},
url = {https://troja.uksw.edu.pl/pdf/kaeiog/KAEiOG2006.365-374.pdf},
issn = {0137-2343},
year = {2006},
date = {2006-01-01},
booktitle = {Evolutionary Computation and Global Optimization},
number = {156},
pages = {365-374},
publisher = {Warsaw University of Technology Publishing House},
organization = {Warsaw University of Technology Publishing House},
series = {Prace Naukowe, Elektronika},
abstract = {This paper presents a formal model of multi-robot interactions based on dynamic game theory. The application of dynamic game theory involves a sequential decision process evolving in (continuous or discrete) time with more than one decision maker (in our case autonomous robot), one or more performance criteria (cost functionals), and possibly having access to different information. The whole range of robot control problems arising from different levels of "cooperation" between robots can be precisely described using different branches of game theory. If the robots have a common goal and one performance objective they act as a team, then team theory is relevant. A noncooperative game refers to the case in which robots have different goals and independent performance objectives. Also another very important issue in action planning i.e. interaction with unknown or partially unknown environment can be viewed as a game against nature.},
keywords = {kaeiog},
pubstate = {published},
tppubtype = {inproceedings}
}
This paper presents a formal model of multi-robot interactions based on dynamic game theory. The application of dynamic game theory involves a sequential decision process evolving in (continuous or discrete) time with more than one decision maker (in our case autonomous robot), one or more performance criteria (cost functionals), and possibly having access to different information. The whole range of robot control problems arising from different levels of "cooperation" between robots can be precisely described using different branches of game theory. If the robots have a common goal and one performance objective they act as a team, then team theory is relevant. A noncooperative game refers to the case in which robots have different goals and independent performance objectives. Also another very important issue in action planning i.e. interaction with unknown or partially unknown environment can be viewed as a game against nature.
22.
Ireneusz Gościniak
Genetic Modification, Artificial Individual, Vaccine and Serum – Genetic Engineering in Evolutionary Systems Proceedings Article
In: Arabas, J. (Ed.): Evolutionary Computation and Global Optimization, pp. 147-154, Warsaw University of Technology Publishing House Warsaw University of Technology Publishing House, 2006, ISSN: 0137-2343.
Abstract | Links | BibTeX | Tagi: kaeiog
@inproceedings{KAEiOG2006.147-154,
title = {Genetic Modification, Artificial Individual, Vaccine and Serum – Genetic Engineering in Evolutionary Systems},
author = {Ireneusz Gościniak},
editor = {J. Arabas},
url = {https://troja.uksw.edu.pl/pdf/kaeiog/KAEiOG2006.147-154.pdf},
issn = {0137-2343},
year = {2006},
date = {2006-01-01},
booktitle = {Evolutionary Computation and Global Optimization},
number = {156},
pages = {147-154},
publisher = {Warsaw University of Technology Publishing House},
organization = {Warsaw University of Technology Publishing House},
series = {Prace Naukowe, Elektronika},
abstract = {The article shows the possibility of genetic engineering implementation in evolutionary systems as: genetic modification, artificial individual, vaccine and serum. In the article the group of immune systems is also proposed. Results of investigation presented in the article suggests that further development of genetic engineering in evolutionary systems will happen, thus raising their efficiency in problem solving.},
keywords = {kaeiog},
pubstate = {published},
tppubtype = {inproceedings}
}
The article shows the possibility of genetic engineering implementation in evolutionary systems as: genetic modification, artificial individual, vaccine and serum. In the article the group of immune systems is also proposed. Results of investigation presented in the article suggests that further development of genetic engineering in evolutionary systems will happen, thus raising their efficiency in problem solving.
23.
Krzysztof Krawiec
Genetic Programming for Primitive-Based Acquisition of Visual Concepts Proceedings Article
In: Arabas, J. (Ed.): Evolutionary Computation and Global Optimization, pp. 211-222, Warsaw University of Technology Publishing House Warsaw University of Technology Publishing House, 2006, ISSN: 0137-2343.
Abstract | Links | BibTeX | Tagi: kaeiog
@inproceedings{KAEiOG2006.211-222,
title = {Genetic Programming for Primitive-Based Acquisition of Visual Concepts},
author = {Krzysztof Krawiec},
editor = {J. Arabas},
url = {https://troja.uksw.edu.pl/pdf/kaeiog/KAEiOG2006.211-222.pdf},
issn = {0137-2343},
year = {2006},
date = {2006-01-01},
booktitle = {Evolutionary Computation and Global Optimization},
number = {156},
pages = {211-222},
publisher = {Warsaw University of Technology Publishing House},
organization = {Warsaw University of Technology Publishing House},
series = {Prace Naukowe, Elektronika},
abstract = {We describe a novel method for acquisition of higher-level visual concepts us-ing GP-based learners that process attributed visual primitives derived from raw raster im-ages. The approach uses an original evaluation scheme: individuals-learners are rewarded for being able to restore the essential features (here: shape) of the visual stimulus. The approach is general and does not require any a priori knowledge about the particular application or target concept to be learned; the only prerequisite is universal knowledge related to interpretation of visual information, encoded in nodes of GP trees. The paper demonstrates the performance of the method on a specific visual task of acquiring the concept of a trian-gle from examples given in a form of raw raster images.},
keywords = {kaeiog},
pubstate = {published},
tppubtype = {inproceedings}
}
We describe a novel method for acquisition of higher-level visual concepts us-ing GP-based learners that process attributed visual primitives derived from raw raster im-ages. The approach uses an original evaluation scheme: individuals-learners are rewarded for being able to restore the essential features (here: shape) of the visual stimulus. The approach is general and does not require any a priori knowledge about the particular application or target concept to be learned; the only prerequisite is universal knowledge related to interpretation of visual information, encoded in nodes of GP trees. The paper demonstrates the performance of the method on a specific visual task of acquiring the concept of a trian-gle from examples given in a form of raw raster images.
24.
Krzysztof Jauernig, Joanna Kołodziej, Michał Stysło
HGSNash Evolutionary Strategy as an Effective Method of Detecting the Nash Equilibria in n-Person Non-Cooperative Games Proceedings Article
In: Arabas, J. (Ed.): Evolutionary Computation and Global Optimization, pp. 171-178, Warsaw University of Technology Publishing House Warsaw University of Technology Publishing House, 2006, ISSN: 0137-2343.
Abstract | Links | BibTeX | Tagi: kaeiog
@inproceedings{KAEiOG2006.171-178,
title = {HGSNash Evolutionary Strategy as an Effective Method of Detecting the Nash Equilibria in n-Person Non-Cooperative Games},
author = {Krzysztof Jauernig and Joanna Kołodziej and Michał Stysło},
editor = {J. Arabas},
url = {https://troja.uksw.edu.pl/pdf/kaeiog/KAEiOG2006.171-178.pdf},
issn = {0137-2343},
year = {2006},
date = {2006-01-01},
booktitle = {Evolutionary Computation and Global Optimization},
number = {156},
pages = {171-178},
publisher = {Warsaw University of Technology Publishing House},
organization = {Warsaw University of Technology Publishing House},
series = {Prace Naukowe, Elektronika},
abstract = {We present a parallel hierarchical evolutionary strategy HGSNash as a new method of detecting the Nash equilibria in n-person non-cooperative games. The problem of finding the equilibrium points is formulated as a global optimization problem. A definition of the strategy and results of some simple numerical experiments are also included.},
keywords = {kaeiog},
pubstate = {published},
tppubtype = {inproceedings}
}
We present a parallel hierarchical evolutionary strategy HGSNash as a new method of detecting the Nash equilibria in n-person non-cooperative games. The problem of finding the equilibrium points is formulated as a global optimization problem. A definition of the strategy and results of some simple numerical experiments are also included.
25.
Anna Kunicka, Halina Kwaśnicka
HYPERMARKET - an evolutionary paths planner Proceedings Article
In: Arabas, J. (Ed.): Evolutionary Computation and Global Optimization, pp. 247-256, Warsaw University of Technology Publishing House Warsaw University of Technology Publishing House, 2006, ISSN: 0137-2343.
Abstract | Links | BibTeX | Tagi: kaeiog
@inproceedings{KAEiOG2006.247-256,
title = {HYPERMARKET - an evolutionary paths planner},
author = {Anna Kunicka and Halina Kwaśnicka},
editor = {J. Arabas},
url = {https://troja.uksw.edu.pl/pdf/kaeiog/KAEiOG2006.247-256.pdf},
issn = {0137-2343},
year = {2006},
date = {2006-01-01},
booktitle = {Evolutionary Computation and Global Optimization},
number = {156},
pages = {247-256},
publisher = {Warsaw University of Technology Publishing House},
organization = {Warsaw University of Technology Publishing House},
series = {Prace Naukowe, Elektronika},
abstract = {The paper presents a new method for path planning. As a real task we assume path planning during shopping in hypermarkets. The method uses a hierarchical approach, where the first algorithm at the low level is responsible for searching a feasible shortest path between two points in the hypermarket, but the second one, at the high level, is developed for searching the best ordering of purchases and the most suitable cash desk. The main goal is to find the shortest path from an entry, via all assumed purchases to the cash desk. In our study, we developed genetic algorithms at the both levels. However, we tested other approaches, namely Ant Colony Optimization and Dijkstratextquoterights algorithm.},
keywords = {kaeiog},
pubstate = {published},
tppubtype = {inproceedings}
}
The paper presents a new method for path planning. As a real task we assume path planning during shopping in hypermarkets. The method uses a hierarchical approach, where the first algorithm at the low level is responsible for searching a feasible shortest path between two points in the hypermarket, but the second one, at the high level, is developed for searching the best ordering of purchases and the most suitable cash desk. The main goal is to find the shortest path from an entry, via all assumed purchases to the cash desk. In our study, we developed genetic algorithms at the both levels. However, we tested other approaches, namely Ant Colony Optimization and Dijkstratextquoterights algorithm.
26.
Iwona Karcz-Dulęba
Identification of Cyclic and Chaotic Behavior in a Dynamical System Generated by Phenotypic Evolution Proceedings Article
In: Arabas, J. (Ed.): Evolutionary Computation and Global Optimization, pp. 195-202, Warsaw University of Technology Publishing House Warsaw University of Technology Publishing House, 2006, ISSN: 0137-2343.
Abstract | Links | BibTeX | Tagi: kaeiog
@inproceedings{KAEiOG2006.195-202,
title = {Identification of Cyclic and Chaotic Behavior in a Dynamical System Generated by Phenotypic Evolution},
author = {Iwona Karcz-Dulęba},
editor = {J. Arabas},
url = {https://troja.uksw.edu.pl/pdf/kaeiog/KAEiOG2006.195-202.pdf},
issn = {0137-2343},
year = {2006},
date = {2006-01-01},
booktitle = {Evolutionary Computation and Global Optimization},
number = {156},
pages = {195-202},
publisher = {Warsaw University of Technology Publishing House},
organization = {Warsaw University of Technology Publishing House},
series = {Prace Naukowe, Elektronika},
abstract = {A discrete deterministic dynamical system generated by the expected value derived from the model of phenotypic evolution is considered. Depending on fitness functions and a standard deviation of mutation, the system converges not only to stable fixed points but also displays cyclic and chaotic behavior. To detect the phenomena an auto-correlation function, a phase space portrait and a power spectrum of trajectories of the system were exploited.},
keywords = {kaeiog},
pubstate = {published},
tppubtype = {inproceedings}
}
A discrete deterministic dynamical system generated by the expected value derived from the model of phenotypic evolution is considered. Depending on fitness functions and a standard deviation of mutation, the system converges not only to stable fixed points but also displays cyclic and chaotic behavior. To detect the phenomena an auto-correlation function, a phase space portrait and a power spectrum of trajectories of the system were exploited.
27.
Michał Bereta, Tadeusz Burczyński
Immune K-Means: A Novel Immune Algorithm for Data Clustering and Multiple-Class Discrimination Proceedings Article
In: Arabas, J. (Ed.): Evolutionary Computation and Global Optimization, pp. 49-60, Warsaw University of Technology Publishing House Warsaw University of Technology Publishing House, 2006, ISSN: 0137-2343.
Abstract | Links | BibTeX | Tagi: kaeiog
@inproceedings{KAEiOG2006.49-60,
title = {Immune K-Means: A Novel Immune Algorithm for Data Clustering and Multiple-Class Discrimination},
author = {Michał Bereta and Tadeusz Burczyński},
editor = {J. Arabas},
url = {https://troja.uksw.edu.pl/pdf/kaeiog/KAEiOG2006.49-60.pdf},
issn = {0137-2343},
year = {2006},
date = {2006-01-01},
booktitle = {Evolutionary Computation and Global Optimization},
number = {156},
pages = {49-60},
publisher = {Warsaw University of Technology Publishing House},
organization = {Warsaw University of Technology Publishing House},
series = {Prace Naukowe, Elektronika},
abstract = {This paper presents a novel approach to data clustering and multiple class classi?cation problems. The proposed method is based on a metaphor derived from immune systems, the clonal selection paradigm. A novel clonal selection algorithm – Immune K-Means, is proposed. The proposed system is able to cluster real valued data efficiently and correctly, dynamically estimating the number of clusters. In classification problems discrimination among classes is based on the k-nearest neighbor method. Two different types of suppression are proposed. They enable the evolution of different populations of lymphocytes well suited to a given problem: clustering or classification. The first type of suppression enables the lymphocytes to discover the data distribution while the second type of suppression focuses the lymphocytes on the classestextquoteright boundaries. Primary results on artificial data and a real-world benchmark dataset (Fishertextquoterights Iris Database) as well as a discussion of the parameters of the algorithm are given.},
keywords = {kaeiog},
pubstate = {published},
tppubtype = {inproceedings}
}
This paper presents a novel approach to data clustering and multiple class classi?cation problems. The proposed method is based on a metaphor derived from immune systems, the clonal selection paradigm. A novel clonal selection algorithm – Immune K-Means, is proposed. The proposed system is able to cluster real valued data efficiently and correctly, dynamically estimating the number of clusters. In classification problems discrimination among classes is based on the k-nearest neighbor method. Two different types of suppression are proposed. They enable the evolution of different populations of lymphocytes well suited to a given problem: clustering or classification. The first type of suppression enables the lymphocytes to discover the data distribution while the second type of suppression focuses the lymphocytes on the classestextquoteright boundaries. Primary results on artificial data and a real-world benchmark dataset (Fishertextquoterights Iris Database) as well as a discussion of the parameters of the algorithm are given.
28.
Krzysztof Adamski, Jarosław Arabas, Łukasz Bartnik, Arkadiusz Chrustowski, Krzysztof Jamróz, Rafał Wardziński
Individualstextquoteright Genealogy and the Population Diversity Proceedings Article
In: Arabas, J. (Ed.): Evolutionary Computation and Global Optimization, pp. 11-20, Warsaw University of Technology Publishing House Warsaw University of Technology Publishing House, 2006, ISSN: 0137-2343.
Abstract | Links | BibTeX | Tagi: kaeiog
@inproceedings{KAEiOG2006.11-20,
title = {Individualstextquoteright Genealogy and the Population Diversity},
author = {Krzysztof Adamski and Jarosław Arabas and Łukasz Bartnik and Arkadiusz Chrustowski and Krzysztof Jamróz and Rafał Wardziński},
editor = {J. Arabas},
url = {https://troja.uksw.edu.pl/pdf/kaeiog/KAEiOG2006.11-20.pdf},
issn = {0137-2343},
year = {2006},
date = {2006-01-01},
booktitle = {Evolutionary Computation and Global Optimization},
number = {156},
pages = {11-20},
publisher = {Warsaw University of Technology Publishing House},
organization = {Warsaw University of Technology Publishing House},
series = {Prace Naukowe, Elektronika},
abstract = {In this paper we study the genotype diversity of the base population. We provide evidence that the diversity is closely related to the genealogy of individuals and to the degree of mutation. We give analytical formulas for the diversity distribution in the space of binary and real vectors. The theoretical results are supported by numerical experiments.},
keywords = {kaeiog},
pubstate = {published},
tppubtype = {inproceedings}
}
In this paper we study the genotype diversity of the base population. We provide evidence that the diversity is closely related to the genealogy of individuals and to the degree of mutation. We give analytical formulas for the diversity distribution in the space of binary and real vectors. The theoretical results are supported by numerical experiments.
29.
Katarzyna Wasielewska, Franciszek Seredyński
Learning Classifier Systems: a way of reinforcement learning based on evolutionary techniques Proceedings Article
In: Arabas, J. (Ed.): Evolutionary Computation and Global Optimization, pp. 385-395, Warsaw University of Technology Publishing House Warsaw University of Technology Publishing House, 2006, ISSN: 0137-2343.
Abstract | Links | BibTeX | Tagi: kaeiog
@inproceedings{KAEiOG2006.385-395,
title = {Learning Classifier Systems: a way of reinforcement learning based on evolutionary techniques},
author = {Katarzyna Wasielewska and Franciszek Seredyński},
editor = {J. Arabas},
url = {https://troja.uksw.edu.pl/pdf/kaeiog/KAEiOG2006.385-395.pdf},
issn = {0137-2343},
year = {2006},
date = {2006-01-01},
booktitle = {Evolutionary Computation and Global Optimization},
number = {156},
pages = {385-395},
publisher = {Warsaw University of Technology Publishing House},
organization = {Warsaw University of Technology Publishing House},
series = {Prace Naukowe, Elektronika},
abstract = {Learning classifier systems (LCSs) are rule-based learning machines in which a reinforcement learning is conducted with use of evolutionary techniques. Currently, they are a subject of intensive study and of interesting applications. In this paper we present a review of LCSs. We give a short history of LCSs and overview current models. We also present some interesting and successful applications of LCSs.},
keywords = {kaeiog},
pubstate = {published},
tppubtype = {inproceedings}
}
Learning classifier systems (LCSs) are rule-based learning machines in which a reinforcement learning is conducted with use of evolutionary techniques. Currently, they are a subject of intensive study and of interesting applications. In this paper we present a review of LCSs. We give a short history of LCSs and overview current models. We also present some interesting and successful applications of LCSs.
30.
Władysław Kamiński, Elwira Tomczak
Linear projection procedure in an evolutionary algorithm Proceedings Article
In: Arabas, J. (Ed.): Evolutionary Computation and Global Optimization, pp. 189-194, Warsaw University of Technology Publishing House Warsaw University of Technology Publishing House, 2006, ISSN: 0137-2343.
Abstract | Links | BibTeX | Tagi: kaeiog
@inproceedings{KAEiOG2006.189-194,
title = {Linear projection procedure in an evolutionary algorithm},
author = {Władysław Kamiński and Elwira Tomczak},
editor = {J. Arabas},
url = {https://troja.uksw.edu.pl/pdf/kaeiog/KAEiOG2006.189-194.pdf},
issn = {0137-2343},
year = {2006},
date = {2006-01-01},
booktitle = {Evolutionary Computation and Global Optimization},
number = {156},
pages = {189-194},
publisher = {Warsaw University of Technology Publishing House},
organization = {Warsaw University of Technology Publishing House},
series = {Prace Naukowe, Elektronika},
abstract = {From the view point of practical applications and the theory of evolutionary algorithms, an important problem is the location of many local extrema showing at the same time the position of a global extremum. To accomplish this task, a new space exploration procedure is proposed using a so-called "linear projection". Additionally, a clustering technique is introduced to the selection procedure. Operation of the method is illustrated by searching the function of two variables with extrema separated by a tableland with a box-type limitation. Each time, irrespective of the number of clusters, calculations enabled location of all three extrema.},
keywords = {kaeiog},
pubstate = {published},
tppubtype = {inproceedings}
}
From the view point of practical applications and the theory of evolutionary algorithms, an important problem is the location of many local extrema showing at the same time the position of a global extremum. To accomplish this task, a new space exploration procedure is proposed using a so-called "linear projection". Additionally, a clustering technique is introduced to the selection procedure. Operation of the method is illustrated by searching the function of two variables with extrema separated by a tableland with a box-type limitation. Each time, irrespective of the number of clusters, calculations enabled location of all three extrema.
31.
Przemysław Miazga
A Meta-Heuristics Algorithm for the Global Optimization of Expensive, Simulator Evaluated, Objective Function Proceedings Article
In: Arabas, J. (Ed.): Evolutionary Computation and Global Optimization, pp. 273-280, Warsaw University of Technology Publishing House Warsaw University of Technology Publishing House, 2006, ISSN: 0137-2343.
Abstract | Links | BibTeX | Tagi: kaeiog
@inproceedings{KAEiOG2006.273-280,
title = {A Meta-Heuristics Algorithm for the Global Optimization of Expensive, Simulator Evaluated, Objective Function},
author = {Przemysław Miazga},
editor = {J. Arabas},
url = {https://troja.uksw.edu.pl/pdf/kaeiog/KAEiOG2006.273-280.pdf},
issn = {0137-2343},
year = {2006},
date = {2006-01-01},
booktitle = {Evolutionary Computation and Global Optimization},
number = {156},
pages = {273-280},
publisher = {Warsaw University of Technology Publishing House},
organization = {Warsaw University of Technology Publishing House},
series = {Prace Naukowe, Elektronika},
abstract = {This paper presents the idea of a meta-heuristics algorithm called Evolutionary Controlled Clustering Algorithm (ECCA) designed for implementation in simulation optimization. The method focuses on localization of function optima neighborhoods. A Evolutionary algorithm (EA) with soft selection and gene injection is used for finding basin of attraction. It operates over nodes of a grid created in a continuous parameter space. ECCA manipulates the grid density as well as the simulation accuracy. Clustered data is used for identification of the basin of attraction. Later, surrogate optimization is applied for local optima search. ECCA was optimized for operation in an uncertain and dynamically changing environment of simulation data. It was tested on the design of the shape of waveguide transition. The computer program can be executed concurrently on a multi-processor machine or on a grid of computers.},
keywords = {kaeiog},
pubstate = {published},
tppubtype = {inproceedings}
}
This paper presents the idea of a meta-heuristics algorithm called Evolutionary Controlled Clustering Algorithm (ECCA) designed for implementation in simulation optimization. The method focuses on localization of function optima neighborhoods. A Evolutionary algorithm (EA) with soft selection and gene injection is used for finding basin of attraction. It operates over nodes of a grid created in a continuous parameter space. ECCA manipulates the grid density as well as the simulation accuracy. Clustered data is used for identification of the basin of attraction. Later, surrogate optimization is applied for local optima search. ECCA was optimized for operation in an uncertain and dynamically changing environment of simulation data. It was tested on the design of the shape of waveguide transition. The computer program can be executed concurrently on a multi-processor machine or on a grid of computers.
32.
Piotr Wąsiewicz, Andrzej Płucienniczak
Molecular Inference Network Experimental Approximation Proceedings Article
In: Arabas, J. (Ed.): Evolutionary Computation and Global Optimization, pp. 405-412, Warsaw University of Technology Publishing House Warsaw University of Technology Publishing House, 2006, ISSN: 0137-2343.
Abstract | Links | BibTeX | Tagi: kaeiog
@inproceedings{KAEiOG2006.405-412,
title = {Molecular Inference Network Experimental Approximation},
author = {Piotr Wąsiewicz and Andrzej Płucienniczak},
editor = {J. Arabas},
url = {https://troja.uksw.edu.pl/pdf/kaeiog/KAEiOG2006.405-412.pdf},
issn = {0137-2343},
year = {2006},
date = {2006-01-01},
booktitle = {Evolutionary Computation and Global Optimization},
number = {156},
pages = {405-412},
publisher = {Warsaw University of Technology Publishing House},
organization = {Warsaw University of Technology Publishing House},
series = {Prace Naukowe, Elektronika},
abstract = {Implementation of the inference system based on DNA chains molec-ular computing is a new paradigm to perform calculations using nanotechnology means. This work presents a new approach to the implementation of inference engines based on DNA. It introduces the subject of inference methods designed to be used with molecular expert systems. The main part of this work includes the concept of the inference engine based on a rule tree specially customized to allow implementation using deoxyribonucleic acid chains. The approach presented al-lows the drawing of inferences based on a variable number of predicates using the most reliable techniques employed in standard operations of genetic engineering. In this approach cross cells are bases of multidimensional DNA structures. An experiment was conducted to confirm the capabilities of the implementation sug-gested. In addition, laboratory evaluation results and perspectives for the further use of the proposed architectural approach are discussed.},
keywords = {kaeiog},
pubstate = {published},
tppubtype = {inproceedings}
}
Implementation of the inference system based on DNA chains molec-ular computing is a new paradigm to perform calculations using nanotechnology means. This work presents a new approach to the implementation of inference engines based on DNA. It introduces the subject of inference methods designed to be used with molecular expert systems. The main part of this work includes the concept of the inference engine based on a rule tree specially customized to allow implementation using deoxyribonucleic acid chains. The approach presented al-lows the drawing of inferences based on a variable number of predicates using the most reliable techniques employed in standard operations of genetic engineering. In this approach cross cells are bases of multidimensional DNA structures. An experiment was conducted to confirm the capabilities of the implementation sug-gested. In addition, laboratory evaluation results and perspectives for the further use of the proposed architectural approach are discussed.
33.
Dariusz Wawrzyniak, Andrzej Obuchowicz
New Approach to Fast and Precise Optimization with Evolutionary Algorithms Proceedings Article
In: Arabas, J. (Ed.): Evolutionary Computation and Global Optimization, pp. 397-404, Warsaw University of Technology Publishing House Warsaw University of Technology Publishing House, 2006, ISSN: 0137-2343.
Abstract | Links | BibTeX | Tagi: kaeiog
@inproceedings{KAEiOG2006.397-404,
title = {New Approach to Fast and Precise Optimization with Evolutionary Algorithms},
author = {Dariusz Wawrzyniak and Andrzej Obuchowicz},
editor = {J. Arabas},
url = {https://troja.uksw.edu.pl/pdf/kaeiog/KAEiOG2006.397-404.pdf},
issn = {0137-2343},
year = {2006},
date = {2006-01-01},
booktitle = {Evolutionary Computation and Global Optimization},
number = {156},
pages = {397-404},
publisher = {Warsaw University of Technology Publishing House},
organization = {Warsaw University of Technology Publishing House},
series = {Prace Naukowe, Elektronika},
abstract = {The paper deals with an evolutionary algorithm which uses new meth-ods to control the range of mutation. In order to significantly increase the effi-ciency of finding the optimum, it discovers and exploits knowledge about the state of population in environment in every generation. It allows to be found a solution both quickly and precisely. Due to the division of population into objects dealing with different functions of optimization, it can simultaneously explore as well as exploit solution space. The algorithm works in such a way that it is unable to undergo a premature convergence. There is no possibility of falling into a trap of local optimum. Therefore, it is possible to increase a selective pressure safely, for example with the help of elitist succession.},
keywords = {kaeiog},
pubstate = {published},
tppubtype = {inproceedings}
}
The paper deals with an evolutionary algorithm which uses new meth-ods to control the range of mutation. In order to significantly increase the effi-ciency of finding the optimum, it discovers and exploits knowledge about the state of population in environment in every generation. It allows to be found a solution both quickly and precisely. Due to the division of population into objects dealing with different functions of optimization, it can simultaneously explore as well as exploit solution space. The algorithm works in such a way that it is unable to undergo a premature convergence. There is no possibility of falling into a trap of local optimum. Therefore, it is possible to increase a selective pressure safely, for example with the help of elitist succession.
34.
Robert Nowak, Andrzej Płucienniczak
Non-Deterministic Finite State Automata Built on DNA Proceedings Article
In: Arabas, J. (Ed.): Evolutionary Computation and Global Optimization, pp. 301-308, Warsaw University of Technology Publishing House Warsaw University of Technology Publishing House, 2006, ISSN: 0137-2343.
Abstract | Links | BibTeX | Tagi: kaeiog
@inproceedings{KAEiOG2006.301-308,
title = {Non-Deterministic Finite State Automata Built on DNA},
author = {Robert Nowak and Andrzej Płucienniczak},
editor = {J. Arabas},
url = {https://troja.uksw.edu.pl/pdf/kaeiog/KAEiOG2006.301-308.pdf},
issn = {0137-2343},
year = {2006},
date = {2006-01-01},
booktitle = {Evolutionary Computation and Global Optimization},
number = {156},
pages = {301-308},
publisher = {Warsaw University of Technology Publishing House},
organization = {Warsaw University of Technology Publishing House},
series = {Prace Naukowe, Elektronika},
abstract = {This paper describes non-deterministic finite-state automaton based on DNA strands. The automaton uses massive parallel processing offered by molecular approach for computation and exhibits a number of advantages over traditional electronic implementations. This device is used to analyze DNA mole-cules, whether they are described by specified regular expression. Presented ideas are confirmed by experiment performed in genetic engineering laboratory.},
keywords = {kaeiog},
pubstate = {published},
tppubtype = {inproceedings}
}
This paper describes non-deterministic finite-state automaton based on DNA strands. The automaton uses massive parallel processing offered by molecular approach for computation and exhibits a number of advantages over traditional electronic implementations. This device is used to analyze DNA mole-cules, whether they are described by specified regular expression. Presented ideas are confirmed by experiment performed in genetic engineering laboratory.
35.
Janusz Orkisz, Artur Obrzut
On some attempts of Evolutionary Algorithms efficiency increase Proceedings Article
In: Arabas, J. (Ed.): Evolutionary Computation and Global Optimization, pp. 321-330, Warsaw University of Technology Publishing House Warsaw University of Technology Publishing House, 2006, ISSN: 0137-2343.
@inproceedings{KAEiOG2006.321-330,
title = {On some attempts of Evolutionary Algorithms efficiency increase},
author = {Janusz Orkisz and Artur Obrzut},
editor = {J. Arabas},
url = {https://troja.uksw.edu.pl/pdf/kaeiog/KAEiOG2006.321-330.pdf},
issn = {0137-2343},
year = {2006},
date = {2006-01-01},
booktitle = {Evolutionary Computation and Global Optimization},
number = {156},
pages = {321-330},
publisher = {Warsaw University of Technology Publishing House},
organization = {Warsaw University of Technology Publishing House},
series = {Prace Naukowe, Elektronika},
keywords = {kaeiog},
pubstate = {published},
tppubtype = {inproceedings}
}
36.
Bartłomiej Jacek Kubica, Krzysztof Malinowski
Optimization of Performance of Queuing Systems With Long-Tailed Service Times Proceedings Article
In: Arabas, J. (Ed.): Evolutionary Computation and Global Optimization, pp. 237-245, Warsaw University of Technology Publishing House Warsaw University of Technology Publishing House, 2006, ISSN: 0137-2343.
Abstract | Links | BibTeX | Tagi: kaeiog
@inproceedings{KAEiOG2006.237-245,
title = {Optimization of Performance of Queuing Systems With Long-Tailed Service Times},
author = {Bartłomiej Jacek Kubica and Krzysztof Malinowski},
editor = {J. Arabas},
url = {https://troja.uksw.edu.pl/pdf/kaeiog/KAEiOG2006.237-245.pdf},
issn = {0137-2343},
year = {2006},
date = {2006-01-01},
booktitle = {Evolutionary Computation and Global Optimization},
number = {156},
pages = {237-245},
publisher = {Warsaw University of Technology Publishing House},
organization = {Warsaw University of Technology Publishing House},
series = {Prace Naukowe, Elektronika},
abstract = {This paper considers the problem of the performance optimization of a ąueuing system with long-tailed seryice time. Due to nonconvexity of the problem global optimization methods have to be used. Interval methods and genetic algo-rithms are considered. Approximation of the Laplace transform of the probability distribution function of the seryice time arises as a subproblem. A new method to do it is developed.},
keywords = {kaeiog},
pubstate = {published},
tppubtype = {inproceedings}
}
This paper considers the problem of the performance optimization of a ąueuing system with long-tailed seryice time. Due to nonconvexity of the problem global optimization methods have to be used. Interval methods and genetic algo-rithms are considered. Approximation of the Laplace transform of the probability distribution function of the seryice time arises as a subproblem. A new method to do it is developed.
37.
Michał Marks, Ewa Niewiadomska-Szynkiewicz
Parallel and Distributed Controlled Random Search Algorithms; Case Study Results Proceedings Article
In: Arabas, J. (Ed.): Evolutionary Computation and Global Optimization, pp. 263-272, Warsaw University of Technology Publishing House Warsaw University of Technology Publishing House, 2006, ISSN: 0137-2343.
Abstract | Links | BibTeX | Tagi: kaeiog
@inproceedings{KAEiOG2006.263-272,
title = {Parallel and Distributed Controlled Random Search Algorithms; Case Study Results},
author = {Michał Marks and Ewa Niewiadomska-Szynkiewicz},
editor = {J. Arabas},
url = {https://troja.uksw.edu.pl/pdf/kaeiog/KAEiOG2006.263-272.pdf},
issn = {0137-2343},
year = {2006},
date = {2006-01-01},
booktitle = {Evolutionary Computation and Global Optimization},
number = {156},
pages = {263-272},
publisher = {Warsaw University of Technology Publishing House},
organization = {Warsaw University of Technology Publishing House},
series = {Prace Naukowe, Elektronika},
abstract = {CRS (Controlled Random Search) algorithms for global optimization are considered. The main objective is to present the advantages of developing the parallel and distributed random search algorithms to search for the global solution. A practical example, application of parallel CRS2, CRS4, CRS6, CRSI algorithms and distributed CRS2 algorithm to calculate the optimal prices of products that are sold in the market, are presented. In the final part of the paper the results of numerical experiments performed on the historical data are described and discussed.},
keywords = {kaeiog},
pubstate = {published},
tppubtype = {inproceedings}
}
CRS (Controlled Random Search) algorithms for global optimization are considered. The main objective is to present the advantages of developing the parallel and distributed random search algorithms to search for the global solution. A practical example, application of parallel CRS2, CRS4, CRS6, CRSI algorithms and distributed CRS2 algorithm to calculate the optimal prices of products that are sold in the market, are presented. In the final part of the paper the results of numerical experiments performed on the historical data are described and discussed.
38.
Adam Długosz, Tadeusz Burczyński, Wacław Kuś
Parallel evolutionary algorithms in shape optimization of heat exchangers under thermomechanical loading Proceedings Article
In: Arabas, J. (Ed.): Evolutionary Computation and Global Optimization, pp. 97-105, Warsaw University of Technology Publishing House Warsaw University of Technology Publishing House, 2006, ISSN: 0137-2343.
Abstract | Links | BibTeX | Tagi: kaeiog
@inproceedings{KAEiOG2006.97-105,
title = {Parallel evolutionary algorithms in shape optimization of heat exchangers under thermomechanical loading},
author = {Adam Długosz and Tadeusz Burczyński and Wacław Kuś},
editor = {J. Arabas},
url = {https://troja.uksw.edu.pl/pdf/kaeiog/KAEiOG2006.97-105.pdf},
issn = {0137-2343},
year = {2006},
date = {2006-01-01},
booktitle = {Evolutionary Computation and Global Optimization},
number = {156},
pages = {97-105},
publisher = {Warsaw University of Technology Publishing House},
organization = {Warsaw University of Technology Publishing House},
series = {Prace Naukowe, Elektronika},
abstract = {The paper is devoted to the shape optimization of the heat exchangers under thermomeehanieal loading. To solve the problem the evolutionary algorithm is used as the optimization techniąue The fitness function is computed by means of the FEM commercial software - MSC MARC/MENTAT. In order to reduce the number of design parameters in evolutionary algorithms the shape of the structure is modelled by Bezier curves. Numerical examples for some shape optimization are included.},
keywords = {kaeiog},
pubstate = {published},
tppubtype = {inproceedings}
}
The paper is devoted to the shape optimization of the heat exchangers under thermomeehanieal loading. To solve the problem the evolutionary algorithm is used as the optimization techniąue The fitness function is computed by means of the FEM commercial software - MSC MARC/MENTAT. In order to reduce the number of design parameters in evolutionary algorithms the shape of the structure is modelled by Bezier curves. Numerical examples for some shape optimization are included.
39.
Jacek Dąbrowski, Marek Kubale
Parallel Tabu Search for Graph Coloring Problem Proceedings Article
In: Arabas, J. (Ed.): Evolutionary Computation and Global Optimization, pp. 87-95, Warsaw University of Technology Publishing House Warsaw University of Technology Publishing House, 2006, ISSN: 0137-2343.
Abstract | Links | BibTeX | Tagi: kaeiog
@inproceedings{KAEiOG2006.87-95,
title = {Parallel Tabu Search for Graph Coloring Problem},
author = {Jacek Dąbrowski and Marek Kubale},
editor = {J. Arabas},
url = {https://troja.uksw.edu.pl/pdf/kaeiog/KAEiOG2006.87-95.pdf},
issn = {0137-2343},
year = {2006},
date = {2006-01-01},
booktitle = {Evolutionary Computation and Global Optimization},
number = {156},
pages = {87-95},
publisher = {Warsaw University of Technology Publishing House},
organization = {Warsaw University of Technology Publishing House},
series = {Prace Naukowe, Elektronika},
abstract = {Tabu search is a simple, yet powerful meta-heuristic based on local search that has been often used to solve combinatorial optimization problems like the graph coloring problem. This paper presents current taxonomy of parallel tabu search algorithms and compares three parallelization techniques applied to Tabucol, a sequential TS algorithm for graph coloring. The experimental results are based on graphs available from the DIMACS benchmark suite.},
keywords = {kaeiog},
pubstate = {published},
tppubtype = {inproceedings}
}
Tabu search is a simple, yet powerful meta-heuristic based on local search that has been often used to solve combinatorial optimization problems like the graph coloring problem. This paper presents current taxonomy of parallel tabu search algorithms and compares three parallelization techniques applied to Tabucol, a sequential TS algorithm for graph coloring. The experimental results are based on graphs available from the DIMACS benchmark suite.
40.
Andrzej Dominik, Zbigniew Walczak, Jacek Wojciechowski
Scale Free Distribution of Nodes in Euclidean Graphs Proceedings Article
In: Arabas, J. (Ed.): Evolutionary Computation and Global Optimization, pp. 107-115, Warsaw University of Technology Publishing House Warsaw University of Technology Publishing House, 2006, ISSN: 0137-2343.
Abstract | Links | BibTeX | Tagi: kaeiog
@inproceedings{KAEiOG2006.107-115,
title = {Scale Free Distribution of Nodes in Euclidean Graphs},
author = {Andrzej Dominik and Zbigniew Walczak and Jacek Wojciechowski},
editor = {J. Arabas},
url = {https://troja.uksw.edu.pl/pdf/kaeiog/KAEiOG2006.107-115.pdf},
issn = {0137-2343},
year = {2006},
date = {2006-01-01},
booktitle = {Evolutionary Computation and Global Optimization},
number = {156},
pages = {107-115},
publisher = {Warsaw University of Technology Publishing House},
organization = {Warsaw University of Technology Publishing House},
series = {Prace Naukowe, Elektronika},
abstract = {In the paper we propose a new model of spatial distribution of nodes in graphs which can be represented in the Euclidean space. Such graphs appear in many areas of computer science, for instance wireless networks design, Traveling Salesman and Vehicle Routing Problems. We show analogies between scale-free and Euclidean graphs. Although the distribution of nodetextquoterights degrees in Euclidean graphs is not scale-free, the spatial distribution of nodetextquoterights follows the power law. We analyze distribution of population density in different continents, propose a model to generate such distributions and provide umerical experiments concerning its quality. Finally, the impact of our model on ditextregisterederent NP-complete problems in Euclidean graphs is analyzed.},
keywords = {kaeiog},
pubstate = {published},
tppubtype = {inproceedings}
}
In the paper we propose a new model of spatial distribution of nodes in graphs which can be represented in the Euclidean space. Such graphs appear in many areas of computer science, for instance wireless networks design, Traveling Salesman and Vehicle Routing Problems. We show analogies between scale-free and Euclidean graphs. Although the distribution of nodetextquoterights degrees in Euclidean graphs is not scale-free, the spatial distribution of nodetextquoterights follows the power law. We analyze distribution of population density in different continents, propose a model to generate such distributions and provide umerical experiments concerning its quality. Finally, the impact of our model on ditextregisterederent NP-complete problems in Euclidean graphs is analyzed.
41.
Jan J. Mulawka
Searching DNA Decision Trees Using Quantum Dots Detection Proceedings Article
In: Arabas, J. (Ed.): Evolutionary Computation and Global Optimization, pp. 281-290, Warsaw University of Technology Publishing House Warsaw University of Technology Publishing House, 2006, ISSN: 0137-2343.
Abstract | Links | BibTeX | Tagi: kaeiog
@inproceedings{KAEiOG2006.281-290,
title = {Searching DNA Decision Trees Using Quantum Dots Detection},
author = {Jan J. Mulawka},
editor = {J. Arabas},
url = {https://troja.uksw.edu.pl/pdf/kaeiog/KAEiOG2006.281-290.pdf},
issn = {0137-2343},
year = {2006},
date = {2006-01-01},
booktitle = {Evolutionary Computation and Global Optimization},
number = {156},
pages = {281-290},
publisher = {Warsaw University of Technology Publishing House},
organization = {Warsaw University of Technology Publishing House},
series = {Prace Naukowe, Elektronika},
abstract = {Concept of self-assembly of DNA molecules may be utilized to implement different algorithms of computing. In particular, this methodology is useful in artificial intelligence because it operates on symbols. As has been shown by the author [7] DNA computing is suitable for searching the decision tree which is encoded by DNA molecules. Thus, the algorithms of artificial intelligence can be performed by this technique. In the contribution we demonstrate that by proper encoding and manipulating with DNA molecules it is possible to implement reasoning procedure by searching the decision tree. The method uses standard genetic engineering operations like hybridization, ligation, and purification. Quantum dot technique is applied to speed up detection of the final output eliminating the slow operation of electrophoresis. The approach presented is very simple and fast because the technique used allows an enormous number of molecules to be labeled, reduces instrument tie-up and improves analysis throughout the process.},
keywords = {kaeiog},
pubstate = {published},
tppubtype = {inproceedings}
}
Concept of self-assembly of DNA molecules may be utilized to implement different algorithms of computing. In particular, this methodology is useful in artificial intelligence because it operates on symbols. As has been shown by the author [7] DNA computing is suitable for searching the decision tree which is encoded by DNA molecules. Thus, the algorithms of artificial intelligence can be performed by this technique. In the contribution we demonstrate that by proper encoding and manipulating with DNA molecules it is possible to implement reasoning procedure by searching the decision tree. The method uses standard genetic engineering operations like hybridization, ligation, and purification. Quantum dot technique is applied to speed up detection of the final output eliminating the slow operation of electrophoresis. The approach presented is very simple and fast because the technique used allows an enormous number of molecules to be labeled, reduces instrument tie-up and improves analysis throughout the process.
42.
Ewa Niewiadomska-Szynkiewicz
Simulation-based Optimization; Methods and Practical Application Proceedings Article
In: Arabas, J. (Ed.): Evolutionary Computation and Global Optimization, pp. 291-300, Warsaw University of Technology Publishing House Warsaw University of Technology Publishing House, 2006, ISSN: 0137-2343.
Abstract | Links | BibTeX | Tagi: kaeiog
@inproceedings{KAEiOG2006.291-300,
title = {Simulation-based Optimization; Methods and Practical Application},
author = {Ewa Niewiadomska-Szynkiewicz},
editor = {J. Arabas},
url = {https://troja.uksw.edu.pl/pdf/kaeiog/KAEiOG2006.291-300.pdf},
issn = {0137-2343},
year = {2006},
date = {2006-01-01},
booktitle = {Evolutionary Computation and Global Optimization},
number = {156},
pages = {291-300},
publisher = {Warsaw University of Technology Publishing House},
organization = {Warsaw University of Technology Publishing House},
series = {Prace Naukowe, Elektronika},
abstract = {The paper is concerned with computational research for complex systems. The simulation-based optimization approach, which is widely used in applied science and engineering, is formulated and discussed. The numerical tech-niques that optimize performance of system by using simulation to evaluate the ob jective value are reviewed. The focus is on random search and metaheuris-tics. The practical example - application of simulation optimization to calculate the optimal decisions for controlling the river-basin reservoir system during flood period is presented and discussed.},
keywords = {kaeiog},
pubstate = {published},
tppubtype = {inproceedings}
}
The paper is concerned with computational research for complex systems. The simulation-based optimization approach, which is widely used in applied science and engineering, is formulated and discussed. The numerical tech-niques that optimize performance of system by using simulation to evaluate the ob jective value are reviewed. The focus is on random search and metaheuris-tics. The practical example - application of simulation optimization to calculate the optimal decisions for controlling the river-basin reservoir system during flood period is presented and discussed.
43.
Krzysztof Patan
Simultaneous Perturbation Stochastic Approximation in Global Optimisation Proceedings Article
In: Arabas, J. (Ed.): Evolutionary Computation and Global Optimization, pp. 343-350, Warsaw University of Technology Publishing House Warsaw University of Technology Publishing House, 2006, ISSN: 0137-2343.
Abstract | Links | BibTeX | Tagi: kaeiog
@inproceedings{KAEiOG2006.343-350,
title = {Simultaneous Perturbation Stochastic Approximation in Global Optimisation},
author = {Krzysztof Patan},
editor = {J. Arabas},
url = {https://troja.uksw.edu.pl/pdf/kaeiog/KAEiOG2006.343-350.pdf},
issn = {0137-2343},
year = {2006},
date = {2006-01-01},
booktitle = {Evolutionary Computation and Global Optimization},
number = {156},
pages = {343-350},
publisher = {Warsaw University of Technology Publishing House},
organization = {Warsaw University of Technology Publishing House},
series = {Prace Naukowe, Elektronika},
abstract = {The paper deals with global property analysis of the Simultaneous Perturbation Stochastic Approximation algorithm. It spite of the fact that the algorithm uses an estimation of a gradient, a global optimisation property can be achieved by using properly shaped sequence ck used for generating trial points. The experiments are performed on a number of test functions showing efficiency of the method. The examined algorithm is also used to train a dynamic neural network. A network under consideration is designed using neuron models with internal dynamics. The identification of an industrial process based on real data shows usefullness of the algorithm.},
keywords = {kaeiog},
pubstate = {published},
tppubtype = {inproceedings}
}
The paper deals with global property analysis of the Simultaneous Perturbation Stochastic Approximation algorithm. It spite of the fact that the algorithm uses an estimation of a gradient, a global optimisation property can be achieved by using properly shaped sequence ck used for generating trial points. The experiments are performed on a number of test functions showing efficiency of the method. The examined algorithm is also used to train a dynamic neural network. A network under consideration is designed using neuron models with internal dynamics. The identification of an industrial process based on real data shows usefullness of the algorithm.
44.
Ewa Figielska
Solving a Flowshop Scheduling Problem with Additional Resource Constraints Proceedings Article
In: Arabas, J. (Ed.): Evolutionary Computation and Global Optimization, pp. 139-146, Warsaw University of Technology Publishing House Warsaw University of Technology Publishing House, 2006, ISSN: 0137-2343.
Abstract | Links | BibTeX | Tagi: kaeiog
@inproceedings{KAEiOG2006.139-146,
title = {Solving a Flowshop Scheduling Problem with Additional Resource Constraints},
author = {Ewa Figielska},
editor = {J. Arabas},
url = {https://troja.uksw.edu.pl/pdf/kaeiog/KAEiOG2006.139-146.pdf},
issn = {0137-2343},
year = {2006},
date = {2006-01-01},
booktitle = {Evolutionary Computation and Global Optimization},
number = {156},
pages = {139-146},
publisher = {Warsaw University of Technology Publishing House},
organization = {Warsaw University of Technology Publishing House},
series = {Prace Naukowe, Elektronika},
abstract = {The paper considers the problem of preemptive scheduling in a two-stage flowshop with parallel unrelated machines at the first stage and a single machine at the second stage. At the first stage jobs require some amounts of additional renewable resources. The objective is the minimization of the makespan. The problem is NP-hard. A heuristic which combines column generation technique with a genetic algorithm is proposed. Several problems with randomly generated parameters are solved. The results indicate that the heuristic is a promising tool for solving the considered problem.},
keywords = {kaeiog},
pubstate = {published},
tppubtype = {inproceedings}
}
The paper considers the problem of preemptive scheduling in a two-stage flowshop with parallel unrelated machines at the first stage and a single machine at the second stage. At the first stage jobs require some amounts of additional renewable resources. The objective is the minimization of the makespan. The problem is NP-hard. A heuristic which combines column generation technique with a genetic algorithm is proposed. Several problems with randomly generated parameters are solved. The results indicate that the heuristic is a promising tool for solving the considered problem.