China
Africa
Explore chapters and articles related to this topic
Parametric Object Detection Using Estimation of Distribution Algorithms
Published in Siddhartha Bhattacharyya, Anirban Mukherjee, Indrajit Pan, Paramartha Dutta, Arup Kumar Bhaumik, Hybrid Intelligent Techniques for Pattern Analysis and Understanding, 2017
Ivan Cruz-Aceves, Jesus Guerrero-Turrubiates, Juan Manuel Sierra-Hernandez
The estimation of distribution algorithms (EDAs) represent an extension to the field of evolutionary computation (EC). EDAs are useful to solve problems in the discrete and continuous domain by using some statistical information of potential solutions, also called individuals [8–10]. Similar to EC techniques, EDAs perform the optimization task using binary encoding and selection operators over a set of potential solutions called population. The main difference regarding classical EC techniques is that EDAs replace the crossover and mutation operators by building probabilistic models at each generation based on global statistical information of the best individuals.
Hybrid Cartesian Genetic Programming Algorithms: A Review
Published in Siddhartha Bhattacharyya, Václav Snášel, Indrajit Pan, Debashis De, Hybrid Computational Intelligence, 2019
Johnathan Melo Neto, Heder S. Bernardino, Helio J.C. Barbosa
Probabilistic Algorithms perform the search space exploration using a probabilistic model of candidate solutions [5]. Most of those methods are called Estimation of Distribution Algorithms (EDAs), and pertain to the evolutionary algorithms class that replaces the variation operators by obtaining a probabilistic model of picked individuals, and sampling it to create novel individuals.
Hybrid evolutionary optimisation with learning for production scheduling: state-of-the-art survey on algorithms and applications
Published in International Journal of Production Research, 2018
Estimation of Distribution Algorithms (EDAs) eschew traditional evolutionary operators, such as mutation and crossover, instead create individuals from probability distributions computed from an existing population. In a sense, EDA offspring are not directly made from selected parents as with other EAs, but are instead created via a layer of indirection manifested in the form of statistical models.