Reducing bias and inefficiency in the selection algorithm
Proceedings of the Second International Conference on Genetic Algorithms on Genetic algorithms and their application
Probabilistic reasoning in intelligent systems: networks of plausible inference
Probabilistic reasoning in intelligent systems: networks of plausible inference
Adaptive Probabilistic Networks with Hidden Variables
Machine Learning - Special issue on learning with probabilistic representations
Sketch-based pruning of a solution space within a formal geometric constraint solver
Artificial Intelligence
Bayesian Networks and Decision Graphs
Bayesian Networks and Decision Graphs
Genetic Algorithms in Search, Optimization and Machine Learning
Genetic Algorithms in Search, Optimization and Machine Learning
Introduction to Bayesian Networks
Introduction to Bayesian Networks
Handbook of Evolutionary Computation
Handbook of Evolutionary Computation
Statistical analysis of the main parameters involved in the designof a genetic algorithm
IEEE Transactions on Systems, Man, and Cybernetics, Part C: Applications and Reviews
A constraint-based system for product design and manufacturing
Robotics and Computer-Integrated Manufacturing
Hi-index | 0.01 |
An interesting problem related to geometric constraint solving is the choice of the "good" solution. The suitability and effectiveness of genetic algorithms applied to this problem has been demonstrated but their performance depends on the values assigned to their control parameters. Although there are recommendations in the specialised technical literature about values for these parameters, their optimal settings depend on the problem at hand. Therefore it would be interesting to define a model that automatically adjusts the values of the evolutive parameters as a function of the geometric problem.This paper proposes a meta-model that generates the recommendations for the right parameter values in genetic algorithms operating as a selector mechanism in constructive geometric constraint solvers. It should be stressed that the proposed model is general and automatic. This means that it is applicable to any context and works without the need for any user supervision.