Title :
Study on evolutionary self-adaptation model-building
Author :
Li, Xiao-ou ; Feng, Huan-Qing
Author_Institution :
Inst. of Biomed. Eng., Univ. of Sci. & Technol. of China, Hefei, China
Abstract :
To overcome the difficulties and drawbacks in building the mathematics model by using the traditional methods, the genetic programming is introduced in this paper, and the model is obtained automatically. This paper presents the improved methods for the generation of the initial population and the way of the fitness measure so as to accelerate the optimization rate and convergence performance of the algorithm. The computational experiments indicate that the reformative algorithm improves the optimization efficiency and the precision of mode.
Keywords :
convergence of numerical methods; genetic algorithms; modelling; probability; average tree creation; constrained self-adaptation fitness; convergence; genetic programming; model-building; optimization; probability; Acceleration; Biomedical engineering; Biomedical measurements; Data analysis; Electronic mail; Genetic programming; Mathematical model; Mathematics; Optimization methods; Shape;
Conference_Titel :
Machine Learning and Cybernetics, 2002. Proceedings. 2002 International Conference on
Print_ISBN :
0-7803-7508-4
DOI :
10.1109/ICMLC.2002.1174378