X-means: Extending K-means with Efficient Estimation of the Number of Clusters
ICML '00 Proceedings of the Seventeenth International Conference on Machine Learning
Linkage identification by non-monotonicity detection for overlapping functions
Evolutionary Computation
The WEKA data mining software: an update
ACM SIGKDD Explorations Newsletter
A surrogate-assisted linkage inference approach in genetic algorithms
Proceedings of the 13th annual conference on Genetic and evolutionary computation
| Hi-index | 0.00 |
We present a genetic algorithm that combines linkage learning, surrogate models and informed operators. Linkage learning aims at measuring and exploiting interdependence of groups of genes. Surrogate models are fitness approximators to ease the task of calculating true fitness values. Informed operators generate, evaluate and rank a set of solutions according to their fitness model to return their most fit solution. Our described approach provides on-line perturbation based linkage learning and informed linkage exploitation with novel, specialized operators. Results of experimental runs on several synthetic fitness function compositions are provided to demonstrate significant improvement of the final result quality compared to a conventional GA setup.