*
[MU]
max
[ TU ]
Figure 5: Pareto front and results from GMP*.
An average level of convergence to the Pareto
set, an maximal level, and the average number of
optimal solutions become worse, when the number
of task, number of nodes, and number of computer
types increase. An average level is 37.7% for the
MGP* versus 35,7% for the AMEA*, if the instance
includes 50 tasks, 4 nodes, 5 computer types and
also 220 binary decision variables.
9 CONCLUDING REMARKS
Genetic programming is relatively new paradigm of
artificial intelligence that can be used for finding
Pareto-optimal solutions. A computer program as
a chromosome is a subject of genetic operators such
as recombination, crossover and mutation. It gives
possibility to represent knowledge that is specific to
the problem in more intelligent way than for the data
structure. A genetic algorithm has been applied for
operating on the population of the computer
procedures written in the Matlab language.
Initial numerical experiments confirm that
feasible, sub-optimal in Pareto sense, task
assignments can be found by genetic programming.
That approach permits for obtaining comparable
quality outcomes to advanced evolutionary
algorithm.
Our future works will focus on testing the other
sets of procedures and terminals to find the Pareto-
optimal task assignments for different criteria and
constraints.
REFERENCES
Balicki, J., 2006. Multicriterion Genetic Programming for
Trajectory Planning of Underwater Vehicle. Int.
Journal of Computer Science and Network Security,
Vol. 6, No. 12, December 30, 1-6.
Balicki, J., 2005. Immune Systems in Multi-criterion
Evolutionary Algorithm for Task Assignments in
Distributed Computer System. Lectures Notes in
Computer Science, Vol. 3528, 51-56.
Bernaschi, M., Castiglione, F., Succi, S., 2006. A High
Performance Simulator of the Immune System. Future
Generation Computer System, Vol. 15, 333-342.
Coello Coello, C. A., Van Veldhuizen, D. A., Lamont,
G.B., 2002.
Evolutionary Algorithms for Solving
Multi-Objective Problems
. Kluwer Academic
Publishers, New York.
Deb, K., 2001.
Multi-Objective Optimization using
Evolutionary Algorithms
, John Wiley & Sons,
Chichester.
Fonseca, C. M., Fleming, P. J., 1995. An Overview of
Evolutionary Algorithms in Multiobjective
Optimisation, Evolutionary Computation, Vol. 3, No.
1, 1-16.
Forrest, S., Perelson, A.S., 1991, Genetic Algorithms and
the Immune System. Lecture Notes in Computer
Science, 320-325.
Jerne, N.K., 1984. Idiotypic Networks and Other
Preconceived Ideas. Immunological Revue, Vol. 79,
5-25.
Kim, J. and Bentley, P. J., 2002.
Immune Memory in the
Dynamic Clonal Selection Algorithm.
Proc. of the First
Int. Conf. on Artificial Immune Systems, Canterbury,
57-65.
Koza J.R., Keane M. A., Streeter M. J., Mydlowec W. ,
Yu J., and Lanza G., 2003. Genetic programming IV.
Routine Human-Competitive Machine Intelligence.
Kluwer Academic Publishers, New York.
Koza, J. R., 1992. Genetic Programming: On the
Programming of Computers by Means of Natural
Selection. Cambridge, MA: The MIT Press.
Samuel A. L., 1960. Programming Computers to Play
Games.
Advances in Computers 1: 165-192.
Sheble, G. B., Britting, K., 1995. Refined Genetic
Algorithm – Economic Dispatch Example. IEEE
Transactions on Power Systems, Vol. 10, No. 2, 117-
124.
Weglarz, J., Nabrzyski, J., Schopf, J., 2003, Grid Resource
Management: State of the Art and Future Trends.
Kluwer Academic Publishers, Boston.
Wierzchon, S. T., 2005. Immune-based Recommender
System. In O. Hryniewicz, J. Kacprzyk, J. Koronacki
and S. T. Wierzchon (eds.) Issues in Intelligent
Systems. Paradigms. Exit, Warsaw, 341-356.
Zitzler, E., Deb, K., and Thiele, L., 2000. Comparison of
Multiobjective Evolutionary Algorithms: Empirical
Results. Evolutionary Computation, Vol. 8, No. 2 173-
195.
MULTI-CRITERION GENETIC PROGRAMMING WITH NEGATIVE SELECTION FOR FINDING PARETO
SOLUTIONS
127