![](bg8.png)
Figure 7: Change of fitness of Santa Fe Trail obtained for
10,000 generations.
and the island model also shows their effectiveness
to improve the fitness.
5 CONCLUSIONS
We applied our methods, GP
CN_CP
, GP
CN_IL
,
GP
CN_CP
(e), and GP
CN_IL
(e) to a garbage collection
problem and the Santa Fe Trail problem, to assess
their performance. In those problems, our methods
show good performance in both the maximum
fitness and the evolution rate. The authors consider
that using conditional probabilities and the island
model prevented the solution from reaching a local
optimum. Additionally, results show that the method
to obtain the optimal value of P improves the fitness.
To improve the fitness of the sub-population of
GP
CN_IL
, our future work will integrate the
conditional probability shown to be effective into
GP
CN_IL
.
This research was in part supported by a
Hiroshima City University Grant for Special
Academic Research (General).
REFERENCES
Koza, J. R., 1992. Genetic Programming: On the
Programming of Computers by Natural Selection,
Cambridge, MA: MIT Press.
Hirasawa, K., Okubo, M., Katagiri, H., Hu, J., and Murata,
J., 2001. Comparison between Genetic Network
Programming and Genetic Programming Using
Evolution of Ant’s Behaviors, IEEJ Transactions on
Electronics, Information and System, Vol.121, No.6,
pp.1001-1009.
Iba, H., 2002. Genetic Algorithm, Igaku Shuppan. Japan.
Mesot, B., Sanchez, E., Pena, C.-A., and Perez-Uribe, A.,
2002. SOS++: Finding Smart Behaviors Using
Learning and Evolution, Eighth International
Conference on the Simulation and Synthesis of Living
Systems (Alife 8), Artificial Life 8, pp.264-273.
Tanji, M., and Iba, H., 2010. A New GP Recombination
Method Using Random Tree Sampling, IEEJ
Transactions on Electronics, Information and Systems,
Vol.130, No.5, pp.775-781.
Iba, H., 2002. Genetic Programming, University of Tokyo
Press.
Murata, T., and Nakamura, T., 2006. Multi-Start Node
Genetic Network Programming for Controlling
Multiple Agents, 2006 IEEE International Conference
on Systems, Man, and Cybernetics, Vol.3, pp.1927-
1932.
Eto, S., Mabu, S., Hirasawa, K., Huruzuki, T., 2007.
Genetic Network Programming with Control Nodes,
2007 IEEE Congress on Evolutionary Computation
(CEC2007), pp.1023-1028.
Minesaki, T., Ueda, H., and Takahashi, K., 2009.
Comparison experiment using Genetic Network
Programming, The Conference Program of the 2009
(60th) Chugoku-branch Joint Convention of Institutes
of Electrical and Information Engineers, p.546.
Morioka, T., Ueda, H., and Takahashi, K., 2011. Efficient
Evolutionary Learning of Agent Behavior by Genetic
Programming Using the Conditional Probabilities,
Proc. of 12
th
International Symposium on Advanced
Intelligent System 2011 (ISIS2011), pp.342-345.
Ito, T., Takahashi, K., Inaba, M., 2013. Improvement of
Genetic Programming with multiple trees, 2013 IEEE
SMC Hiroshima Chapter Young Researchers'
Workshop Proceedings, pp.9-12.
Ono, K., Hanada, Y., Shirakawa, K., Kumano, M., Kimura,
M., 2012. Depth-dependent crossover in genetic
programming with frequent trees, 2012 IEEE
International Conference on Systems, Man, and
Cybernetics, pp.359-363.
Ono, K., Hanada, Y., Kumano, M., Kimura, M., 2013.
Island model genetic programming based on frequent
trees, 2013 IEEE Conference on Evolutionary
Computation (CEC2013), pp.2988-2995.
Iwashita, M., and Iba, H., 2002. Parallel Distributed GP
with Immigrants Aging and Depth-dependent
Crossover, Transactions of Information Processing
Society of Japan, Vol.43, No.SIG10, pp.146-156.
0
10
20
30
40
50
60
70
80
0 2000 4000 6000 8000 10000
GNP_CN GP_CN
GP_Cne GP_CNCP
GP_CNCPe GP_CNIL
GP_CNILe
Fitness
Generation
GNP
CN
GP
CN
(e)
GP
CN_CP
(e)
GP
CN_IL
(e)
GP
CN
GP
CN_CP
GP
CN_IL
ExperimentsAssessingLearningofAgentBehaviorusingGeneticProgrammingwithMultipleTrees
271