Fitness Value Curves Prediction in the Evolutionary Process of Genetic Algorithms Applied to Benchmark Function
Resumo
This work intends to adopt fitness curves prediction from Genetic Algorithms (GAs) proposed in [Almeida et al. 2021], in the context of a more complex function, which is the Schwefel benchmark function. The prediction is performed with the knowledge only of the GA initialization parameters, using the Random Forest model. This approach addresses the main gap in the original work achieving good results, which makes this approach more promising.
Referências
Bhattacharya, M. (2013). Evolutionary approaches to expensive optimisation. International Journal of Advanced Research in Artificial Intelligence, 2(3).
Cutler, A., Cutler, D. R., and Stevens, J. R. (2012). Random forests. In Ensemble machine learning, pages 157–175. Springer.
Eiben, A. E. and Smit, S. K. (2012). Evolutionary algorithm parameters and methods to tune them. In Hamadi, Y., Monfroy, E., and Saubion, F., editors, Autonomous Search, pages 15–36. Springer.
Goldberg, D. E. (1989). Genetic algorithms in search. Optimization, and MachineLearning.
Holland, J. H. et al. (1992). Adaptation in natural and artificial systems: an introductory analysis with applications to biology, control, and artificial intelligence. MIT press.
Hussain, K., Salleh, M., Cheng, S., and Naseem, R. (2017). Common benchmark functions for metaheuristic evaluation: A review. International Journal on Informatics Visualization, 1:218–223.
Mitchell, M. (1998). An Introduction to Genetic Algorithms. MIT Press, Cambridge, MA, USA.