Category: Evolutionary Intelligence

Surrogate-assisted clonal selection algorithms for expensive optimization problems

Abstract  Clonal selection algorithms are computational methods inspired by the behavior of the immune system which can be applied to
solve optimization problems. However, like other nature inspired algorithms, they can require a large numbe…

Abstract  

Clonal selection algorithms are computational methods inspired by the behavior of the immune system which can be applied to
solve optimization problems. However, like other nature inspired algorithms, they can require a large number of objective
function evaluations in order to reach a satisfactory solution. When those evaluations involve a computationally expensive
simulation model their cost becomes prohibitive. In this paper we analyze the use of surrogate models in order to enhance
the performance of a clonal selection algorithm. Computational experiments are conducted to assess the performance of the
presented techniques using a benchmark with 22 test-problems under a fixed budget of objective function evaluations. The comparisons
show that for most cases the use of surrogate models improve significantly the performance of the baseline clonal selection
algorithm.

  • Content Type Journal Article
  • Pages 81-97
  • DOI 10.1007/s12065-011-0056-1
  • Authors
    • Heder S. Bernardino, Laboratório Nacional de Computação Científica, LNCC, Av. Getulio Vargas, 333, 25651-075 Petrópolis, RJ, Brazil
    • Helio J. C. Barbosa, Laboratório Nacional de Computação Científica, LNCC, Av. Getulio Vargas, 333, 25651-075 Petrópolis, RJ, Brazil
    • Leonardo G. Fonseca, Universidade Federal de Juiz de Fora, UFJF, Campus Universitrio, 36036-330 Juiz de Fora, MG, Brazil

Learning classifier systems to evolve classification rules for systems of memory constrained components

Abstract  In this paper we study how to solve classification problems in computing systems that consist of distributed, memory constrained
components. Interacting Pittsburgh-style Learning Classifier Systems are used to generate sets of clas…

Abstract  

In this paper we study how to solve classification problems in computing systems that consist of distributed, memory constrained
components. Interacting Pittsburgh-style Learning Classifier Systems are used to generate sets of classification rules that
can be deployed on the components. We show that this approach distributes the knowledge and enables the components to solve
complex classification problems in cooperation. We study the structure and properties of the evolved rule sets and analyse
the way the components share their knowledge. Moreover, we investigate the influence of different communication topologies
and the introduction of communication costs on the emerging patterns of cooperation and on the classification performance
of the whole system.

  • Content Type Journal Article
  • Category Research Paper
  • Pages 127-143
  • DOI 10.1007/s12065-011-0053-4
  • Authors
    • Alexander Scheidler, Institut de Recherches Interdisciplinaires et de Développements en Intelligence Artificielle Université Libre de Bruxelles IRIDIA, CP 194/6, 1050 Bruxelles, Belgium
    • Martin Middendorf, Department of Computer Science of University of Leipzig, Johannisgasse 26, 04103 Leipzig, Germany

On the role of age diversity for effective aging operators

Abstract  Aging is a general mechanism that some randomized search heuristics employ to increase the diversity of their collection of
search points. A more diverse collection of search points is believed to improve the search heuristic’s p…

Abstract  

Aging is a general mechanism that some randomized search heuristics employ to increase the diversity of their collection of
search points. A more diverse collection of search points is believed to improve the search heuristic’s performance for difficult
problems. The most prominent randomized search heuristics with aging are evolutionary algorithms and artificial immune systems.
While it is known that randomized search heuristics with aging can be very much more efficient than randomized search heuristics
without aging the details of the origin of such benefits are difficult to understand. We contribute to this understanding
by presenting a detailed and structured analysis of aging. We prove that in addition to diversity with respect to search points
diversity with respect to age plays a key role. We analyze different ways of dealing with age diversity by means of theoretical
as well as empirical analyses. Major results include a more structured understanding of aging and showcases where age diversity
can make the difference between efficient and completely inefficient optimization.

  • Content Type Journal Article
  • Pages 99-125
  • DOI 10.1007/s12065-011-0051-6
  • Authors
    • Thomas Jansen, Department of Computer Science, University College Cork, Cork, Ireland
    • Christine Zarges, Lehrstuhl 2, Fakultät für Informatik, TU Dortmund, 44221 Dortmund, Germany

Collective classification of textual documents by guided self-organization in T-Cell cross-regulation dynamics

Abstract  We present and study an agent-based model of T-Cell cross-regulation in the adaptive immune system, which we apply to binary
classification. Our method expands an existing analytical model of T-cell cross-regulation (Carneiro et al…

Abstract  

We present and study an agent-based model of T-Cell cross-regulation in the adaptive immune system, which we apply to binary
classification. Our method expands an existing analytical model of T-cell cross-regulation (Carneiro et al. in Immunol Rev
216(1):48–68, 2007) that was used to study the self-organizing dynamics of a single population of T-Cells in interaction with an idealized antigen
presenting cell capable of presenting a single antigen. With agent-based modeling we are able to study the self-organizing
dynamics of multiple populations of distinct T-cells which interact via antigen presenting cells that present hundreds of
distinct antigens. Moreover, we show that such self-organizing dynamics can be guided to produce an effective binary classification
of antigens, which is competitive with existing machine learning methods when applied to biomedical text classification. More
specifically, here we test our model on a dataset of publicly available full-text biomedical articles provided by the BioCreative
challenge (Krallinger in The biocreative ii. 5 challenge overview, p 19, 2009). We study the robustness of our model’s parameter configurations, and show that it leads to encouraging results comparable
to state-of-the-art classifiers. Our results help us understand both T-cell cross-regulation as a general principle of guided
self-organization, as well as its applicability to document classification. Therefore, we show that our bio-inspired algorithm
is a promising novel method for biomedical article classification and for binary document classification in general.

  • Content Type Journal Article
  • Pages 69-80
  • DOI 10.1007/s12065-011-0052-5
  • Authors
    • Alaa Abi-Haidar, School of Informatics and Computing, Indiana University, Bloomington, IN 47401, USA
    • Luis M. Rocha, School of Informatics and Computing, Indiana University, Bloomington, IN 47401, USA

Studying the application of ant colony optimization and river formation dynamics to the steiner tree problem

Abstract  
River formation dynamics (RFD) (Rabanal et al. in Using river formation dynamics to design heuristic algorithms. In: Unconventional computation, UC’07,
LNCS 4618. Springer, pp 163–177, 2007; Rabanal et al. in Applyi…

Abstract  

River formation dynamics (RFD) (Rabanal et al. in Using river formation dynamics to design heuristic algorithms. In: Unconventional computation, UC’07,
LNCS 4618. Springer, pp 163–177, 2007; Rabanal et al. in Applying river formation dynamics to solve NP-complete problems. In: Chiong R (ed) Nature-inspired algorithms
for optimisation, volume 193 of Studies in Computational Intelligence. Springer, pp 333–368, 2009) is a heuristic optimization algorithm based on copying how water forms rivers by eroding the ground and depositing sediments.
We apply this method to solve the Steiner tree problem (STP), a well-known NP-hard problem having applications to areas like telecommunications routing or VLSI design among many
others. We show that the gradient orientation of RFD makes it especially suitable for solving this problem, and we report
the results of several experiments where RFD, as well as an implementation of Ant Colony Optimization (ACO) (Dorigo in ant colony optimization. MIT Press, New York, 2004), are applied to some benchmark graphs from the SteinLib Testdata Library (Koch in Steinlib testdata library. Technical report,
Konrad-Zuse-Zentrum für Informationstechnik Berlin, 2009). We also study the capability of RFD and ACO to deal with a scenario where the graph is modified after a solution is found, and next a solution for the new graph has to be found – either by running the algorithm from scratch
or by adapting the structures previously formed by the algorithms to construct their previous solution.

  • Content Type Journal Article
  • Pages 51-65
  • DOI 10.1007/s12065-011-0049-0
  • Authors
    • Pablo Rabanal, Dept. Sistemas Informáticos y Computación. Facultad de Informática, Universidad Complutense de Madrid, C/Profesor José García Santesmases, s/n, 28040 Madrid, Spain
    • Ismael Rodríguez, Dept. Sistemas Informáticos y Computación. Facultad de Informática, Universidad Complutense de Madrid, C/Profesor José García Santesmases, s/n, 28040 Madrid, Spain
    • Fernando Rubio, Dept. Sistemas Informáticos y Computación. Facultad de Informática, Universidad Complutense de Madrid, C/Profesor José García Santesmases, s/n, 28040 Madrid, Spain

A hybrid harmony search algorithm for MRI brain segmentation

Abstract  Automatic magnetic resonance imaging (MRI) brain segmentation is a challenging problem that has received significant attention
in the field of medical image processing. In this paper, we present a new dynamic clustering algorithm b…

Abstract  

Automatic magnetic resonance imaging (MRI) brain segmentation is a challenging problem that has received significant attention
in the field of medical image processing. In this paper, we present a new dynamic clustering algorithm based on the hybridization
of harmony search (HS) and fuzzy c-means to automatically segment MRI brain images in an intelligent manner. In our algorithm,
the capability of standard HS is modified to automatically evolve the appropriate number of clusters as well as the locations
of cluster centers. By incorporating the concept of variable length encoding in each harmony memory vector, this algorithm
is able to represent variable numbers of candidate cluster centers at each iteration. A new HS operator, called the “empty operator”, has been introduced to support the selection of empty decision variables in the harmony memory vector. The PBMF cluster
validity index is used as an objective function to validate the clustering result obtained from each harmony memory vector.
Evaluation of the proposed algorithm has been performed using both real MRI data obtained from the Center for Morphometric
Analysis at Massachusetts General Hospital and simulated MRI data generated using the McGill University BrainWeb MRI simulator.
Experimental results show the ability of this algorithm to find the appropriate number of naturally occurring regions in brain
images. Furthermore, the superiority of the proposed algorithm over various state-of-the-art segmentation algorithms is demonstrated
quantitatively.

  • Content Type Journal Article
  • Pages 31-49
  • DOI 10.1007/s12065-011-0048-1
  • Authors
    • Osama Moh’d Alia, Computer Vision Research Group, School of Computer Sciences, University Sains Malaysia, 11800 USM Penang, Malaysia
    • Rajeswari Mandava, Computer Vision Research Group, School of Computer Sciences, University Sains Malaysia, 11800 USM Penang, Malaysia
    • Mohd Ezane Aziz, Department of Radiology-Health Campus, Universiti Sains Malaysia, 16150 Kubang Kerian, Kelantan Malaysia

Special issue on modern search heuristics and applications

Special issue on modern search heuristics and applications
Content Type Journal ArticlePages 1-2DOI 10.1007/s12065-011-0050-7Authors
Raymond Chiong, Faculty of Information & Communication Technologies, Swinburne University of Technology, Melbourne, …

Special issue on modern search heuristics and applications

  • Content Type Journal Article
  • Pages 1-2
  • DOI 10.1007/s12065-011-0050-7
  • Authors
    • Raymond Chiong, Faculty of Information & Communication Technologies, Swinburne University of Technology, Melbourne, VIC Australia
    • Thomas Weise, Nature Inspired Computation & Applications Laboratory, University of Science and Technology of China, Hefei, Anhui China

Novel evolutionary algorithms for supervised classification problems: an experimental study

Abstract  Evolutionary Algorithms (EAs) are population-based, stochastic search algorithms that mimic natural evolution. Over the years,
EAs have been successfully applied to many classification problems. In this paper, we present three nove…

Abstract  

Evolutionary Algorithms (EAs) are population-based, stochastic search algorithms that mimic natural evolution. Over the years,
EAs have been successfully applied to many classification problems. In this paper, we present three novel evolutionary approaches
and analyze their performances for synthesizing classifiers with EAs in supervised data mining scenarios. The first approach
is based on encoding rule sets with bit string genomes, while the second one utilizes Genetic Programming (GP) to create decision
trees with arbitrary expressions attached to the nodes. The novelty of these two approaches lies in the use of solutions on
the Pareto front as an ensemble. The third approach, EDDIE-101, is also based on GP but uses a new, advanced fitness measure
and some novel genetic operators. We compare these approaches to a number of well-known data mining methods, including C4.5
and Random-Forest, and show that the performances of our evolved classifiers can be very competitive as far as the solution
quality is concerned. In addition, the proposed approaches work well across a wide range of configurations, and EDDIE-101
particularly has been highly efficient. To further evaluate the flexibility of EDDIE-101 across different problem domains,
we also test it on some real financial datasets for finding investment opportunities and compare the results with those obtained
using other classifiers. Numerical experiments confirm that EDDIE-101 can be successfully extended to financial forecasting.

  • Content Type Journal Article
  • Pages 3-16
  • DOI 10.1007/s12065-010-0047-7
  • Authors
    • Pu Wang, Nature Inspired Computation & Applications Laboratory, University of Science and Technology of China, Hefei, Anhui China
    • Thomas Weise, Nature Inspired Computation & Applications Laboratory, University of Science and Technology of China, Hefei, Anhui China
    • Raymond Chiong, Faculty of Information & Communication Technologies, Swinburne University of Technology, Melbourne, VIC 3122, Australia

Composed compact differential evolution

Abstract  This paper proposes a novel algorithm for solving continuous complex optimization problems with a relatively low memory consumption.
The proposed approach, namely Composed compact Differential Evolution, consists of a set of compac…

Abstract  

This paper proposes a novel algorithm for solving continuous complex optimization problems with a relatively low memory consumption.
The proposed approach, namely Composed compact Differential Evolution, consists of a set of compact Differential Evolution
units which simultaneously search the decision space from various perspectives. A randomization in the virtual population
allows the algorithm to behave, on one hand, as a multiple local search with a multi-start logic integrated within it. On
the other hand, the compact units communicate among each other by means of a ring topology and propagation of information.
More specifically, the most promising elite solutions and scale factor values of each compact unit are migrated to the neighbour
unit so that the search of the global optimum is performed. In other words, while each single compact unit performs a local
search by exploiting the direction suggested by each elite solution, the entire structure combines the achievement of each
local search operation towards the direction of the global search. The proposed algorithm is characterized by a limited memory
consumption and is memory-wise equivalent to a population-based algorithm with a small population. Numerical results show
that the proposed approach outperforms other compact algorithms and various modern population-based structures.

  • Content Type Journal Article
  • Pages 17-29
  • DOI 10.1007/s12065-010-0046-8
  • Authors
    • Giovanni Iacca, Department of Mathematical Information Technology, University of Jyväskylä, P.O. Box 35, Agora, 40014 Finland
    • Ernesto Mininno, Department of Mathematical Information Technology, University of Jyväskylä, P.O. Box 35, Agora, 40014 Finland
    • Ferrante Neri, Department of Mathematical Information Technology, University of Jyväskylä, P.O. Box 35, Agora, 40014 Finland

Memetic pareto differential evolutionary artificial neural networks to determine growth multi-classes in predictive microbiology

Abstract  The main objective of this research is to automatically design Artificial Neural Network models with sigmoid basis units for
multiclassification tasks in predictive microbiology. The classifiers obtained achieve a double objective:…

Abstract  

The main objective of this research is to automatically design Artificial Neural Network models with sigmoid basis units for
multiclassification tasks in predictive microbiology. The classifiers obtained achieve a double objective: a high classification
level in the dataset and high classification levels for each class. The Memetic Pareto Differential Evolution Neural Network
chosen to learn the structure and weights of the Neural Networks is a Differential Evolutionary approach based on the Pareto
Differential Evolution multiobjective evolutionary algorithm. The Pareto Differential Evolution algorithm is augmented with
a local search using the improved Resilient Backpropagation with backtracking–iRprop
+ algorithm. To analyze the robustness of this methodology, it has been applied to two complex classification problems in predictive
microbiology (Staphylococcus aureus and Shigella flexneri). The results obtained show that the generalization ability and the classification rate in each class can be more efficiently
improved within this multiobjective algorithm.

  • Content Type Journal Article
  • DOI 10.1007/s12065-010-0045-9
  • Authors
    • M. Cruz-Ramírez, Department of Computer Science and Numerical Analysis, University of Córdoba, Rabanales Campus, Albert Einstein building 3 floor, 14071 Córdoba, Spain
    • J. Sánchez-Monedero, Department of Computer Science and Numerical Analysis, University of Córdoba, Rabanales Campus, Albert Einstein building 3 floor, 14071 Córdoba, Spain
    • F. Fernández-Navarro, Department of Computer Science and Numerical Analysis, University of Córdoba, Rabanales Campus, Albert Einstein building 3 floor, 14071 Córdoba, Spain
    • J. C. Fernández, Department of Computer Science and Numerical Analysis, University of Córdoba, Rabanales Campus, Albert Einstein building 3 floor, 14071 Córdoba, Spain
    • C. Hervás-Martínez, Department of Computer Science and Numerical Analysis, University of Córdoba, Rabanales Campus, Albert Einstein building 3 floor, 14071 Córdoba, Spain