A Quantitative Comparison Between MOGAs and the RRT Algorithm on Classification Systems Optimization
Resumo
Algoritmos genéticos são métodos de otimização baseados em população. Seus equivalentes multi-critério são usados freqüentemente na otimização de sistemas de classificação, mas pouco se discute sobre o custo computacional ao solucionar tais problemas. Para entender melhor esta relação, é proposta a utilização de uma abordagem baseada em simulated annealing. Os resultados são comparados com os obtidos por algoritmos genéticos multi-critério no mesmo problema. Os experimentos com dígitos manuscritos isolados indicam a eficácia e baixo custo computacional da abordagem baseada em simulated annealing.Referências
Handl, J. and Knowles, J. (2006). Feature subset selection in unsupervised learning via multiobjective optimization. International Journal on Computational Intelligence Research, 3(1):217–238.
Kimura, F., Inoue, S., Wakabayashi, T., Tsuruoka, S., and Miyake, Y. (1998). Handwritten Numeral Recognition using Autoassociative Neural Networks. In Proceedings of the International Conference on Pattern Recognition, pages 152–155.
Kittler, J., Hatef, M., Duin, R. P. W., and Matas, J. (1998). On combining classifiers. IEEE Transactions on Pattern Analysis and Machine Intelligence, 20(3):226–239.
Kuncheva, L. I. and Jain, L. C. (2000). Design classifier fusion systems by genetic algorithms. IEEE Transactions on Evolutionary Computation, 4(4):327–336.
Li, Z.-C. and Suen, C. Y. (2000). The partition-combination method for recognition of handwritten characters. Pattern Recognition Letters, 21(8):701–720.
Oliveira, L. S., Sabourin, R., Bortolozzi, F., and Suen, C. Y. (2002). Automatic Recognition of Handwritten Numerical Strings: A Recognition and Verification Strategy. IEEE Trans. on Pattern Analysis and Machine Intelligence, 24(11):1438–1454.
Pepper, J. W., Golden, B. L., and Wasil, E. A. (2002). Solving the traveling salesman problem with annealing-based heuristics: A computational study. IEEE Trans. on Systems, Mand and Cybernetics – Part A: Systems and Humans, 32(1):72–77.
Radtke, P. V. W., Wong, T., and Sabourin, R. (2006a). Classification system optimization with multi-objective genetic algorithms. In Proceedings of the 10th International Workshop on Frontiers in Handwriten Recognition (IWFHR 2006), pages 331–336. IAPR.
Radtke, P. V. W., Wong, T., and Sabourin, R. (2006b). An evaluation of over-fit control strategies for multi-objective evolutionary optimization. In Proceedings of the International Joint Conference on Neural Networks (IJCNN 2006), pages 6359–6366. IEEE Computer Society.
Ruta, D. and Gabrys, B. (2005). Classifier Selection for Majority Voting. Information fusion, 6:63–81.
Tremblay, G., Sabourin, R., and Maupin, P. (2004). Optimizing nearest neighbour in random subspaces using a multi-objective genetic algorithm. In 17th International Conference on Pattern Recognition – ICPR2004, pages 208–211, Cambridge, U.K. IEEE Computer Society.
Tsymbal, A., Pechenizkiy, M., and Cunningham, P. (2005). Sequential genetic search for ensemble feature selection. In Proceddings of International Joint Conference on Artificial Intelligence, pages 877–882.
Kimura, F., Inoue, S., Wakabayashi, T., Tsuruoka, S., and Miyake, Y. (1998). Handwritten Numeral Recognition using Autoassociative Neural Networks. In Proceedings of the International Conference on Pattern Recognition, pages 152–155.
Kittler, J., Hatef, M., Duin, R. P. W., and Matas, J. (1998). On combining classifiers. IEEE Transactions on Pattern Analysis and Machine Intelligence, 20(3):226–239.
Kuncheva, L. I. and Jain, L. C. (2000). Design classifier fusion systems by genetic algorithms. IEEE Transactions on Evolutionary Computation, 4(4):327–336.
Li, Z.-C. and Suen, C. Y. (2000). The partition-combination method for recognition of handwritten characters. Pattern Recognition Letters, 21(8):701–720.
Oliveira, L. S., Sabourin, R., Bortolozzi, F., and Suen, C. Y. (2002). Automatic Recognition of Handwritten Numerical Strings: A Recognition and Verification Strategy. IEEE Trans. on Pattern Analysis and Machine Intelligence, 24(11):1438–1454.
Pepper, J. W., Golden, B. L., and Wasil, E. A. (2002). Solving the traveling salesman problem with annealing-based heuristics: A computational study. IEEE Trans. on Systems, Mand and Cybernetics – Part A: Systems and Humans, 32(1):72–77.
Radtke, P. V. W., Wong, T., and Sabourin, R. (2006a). Classification system optimization with multi-objective genetic algorithms. In Proceedings of the 10th International Workshop on Frontiers in Handwriten Recognition (IWFHR 2006), pages 331–336. IAPR.
Radtke, P. V. W., Wong, T., and Sabourin, R. (2006b). An evaluation of over-fit control strategies for multi-objective evolutionary optimization. In Proceedings of the International Joint Conference on Neural Networks (IJCNN 2006), pages 6359–6366. IEEE Computer Society.
Ruta, D. and Gabrys, B. (2005). Classifier Selection for Majority Voting. Information fusion, 6:63–81.
Tremblay, G., Sabourin, R., and Maupin, P. (2004). Optimizing nearest neighbour in random subspaces using a multi-objective genetic algorithm. In 17th International Conference on Pattern Recognition – ICPR2004, pages 208–211, Cambridge, U.K. IEEE Computer Society.
Tsymbal, A., Pechenizkiy, M., and Cunningham, P. (2005). Sequential genetic search for ensemble feature selection. In Proceddings of International Joint Conference on Artificial Intelligence, pages 877–882.
Publicado
30/06/2007
Como Citar
RADTKE, Paulo V. W.; SABOURIN, Robert; WONG, Tony.
A Quantitative Comparison Between MOGAs and the RRT Algorithm on Classification Systems Optimization. In: ENCONTRO NACIONAL DE INTELIGÊNCIA ARTIFICIAL E COMPUTACIONAL (ENIAC), 6. , 2007, Rio de Janeiro/RJ.
Anais [...].
Porto Alegre: Sociedade Brasileira de Computação,
2007
.
p. 1460-1468.
ISSN 2763-9061.
