When Will a Genetic Algorithm Outperform Hill Climbing?
Proceedings of the 5th International Conference on Genetic Algorithms
Linkage learning, overlapping building blocks, and systematic strategy for scalable recombination
GECCO '05 Proceedings of the 7th annual conference on Genetic and evolutionary computation
Conquering hierarchical difficulty by explicit chunking: substructural chromosome compression
Proceedings of the 8th annual conference on Genetic and evolutionary computation
Towards billion-bit optimization via a parallel estimation of distribution algorithm
Proceedings of the 9th annual conference on Genetic and evolutionary computation
Sporadic model building for efficiency enhancement of the hierarchical BOA
Genetic Programming and Evolvable Machines
Proceedings of the 10th annual conference on Genetic and evolutionary computation
iBOA: the incremental bayesian optimization algorithm
Proceedings of the 10th annual conference on Genetic and evolutionary computation
Enhancing the Efficiency of the ECGA
Proceedings of the 10th international conference on Parallel Problem Solving from Nature: PPSN X
Correlation guided model building
Proceedings of the 11th Annual conference on Genetic and evolutionary computation
GECCO'03 Proceedings of the 2003 international conference on Genetic and evolutionary computation: PartII
Hierarchical BOA, cluster exact approximation, and ising spin glasses
PPSN'06 Proceedings of the 9th international conference on Parallel Problem Solving from Nature
Model complexity vs. performance in the bayesian optimization algorithm
PPSN'06 Proceedings of the 9th international conference on Parallel Problem Solving from Nature
Hi-index | 0.00 |
Probabilistic models of high-order statistics, capable of expressing complex variable interactions, have been successfully applied by estimation of distribution algorithms (EDAs) to render hard problems tractable. Unfortunately, the dependence structure induction stage in these methods imposes a high computational cost that often dominates the overall complexity of the whole search process. In this paper, a new unsupervised model induction strategy built upon a maximum flow graph clustering technique is presented. The new approach offers a model evaluation free, fast, scalable, easily parallelizable method, capable of complex dependence structure induction. The method can be used to infer different classes of probabilistic models.