Estimation of Distribution algorithms (EDAs) are a class of evolutionary algorithms that use machine learning techniques to solve optimization problems. Machine learning is used to learn probabilistic models of the se...
详细信息
ISBN:
(纸本)9781595931863
Estimation of Distribution algorithms (EDAs) are a class of evolutionary algorithms that use machine learning techniques to solve optimization problems. Machine learning is used to learn probabilistic models of the selected population. This model is then used to generate next population via sampling. An important phenomenon in machine learning from data is called overfitting. This occurs when the model is overly adapted to the specifics of the training data so well that even noise is encoded. The purpose of this paper is to investigate. whether overfitting happens in EDAs, and to discover its consequences. What is found is: overfitting does occur in EDAs;overfitting correlates to EDAs performance;reduction of overfitting using early stopping can improve EDAs performance.
This paper describes and analyzes sporadic model building, which can be used to enhance the efficiency of the hierarchical bayesian optimization algorithm (hBOA) and other advanced estimation of distribution algorithm...
详细信息
ISBN:
(纸本)9781595931863
This paper describes and analyzes sporadic model building, which can be used to enhance the efficiency of the hierarchical bayesian optimization algorithm (hBOA) and other advanced estimation of distribution algorithms (EDAs) that use complex multivariate probabilistic models. With sporadic model building, the structure of the probabilistic model is updated once every few iterations (generations), whereas in the remaining iterations only model parameters (conditional and marginal probabilities) axe updated. Since the time complexity of updating model parameters is much lower than the time complexity of learning the model structure, sporadic model building decreases the overall time complexity of model building. The paper shows that for boundedly difficult nearly decomposable and hierarchical optimization problems, sporadic model building leads to a significant model-building speedup that decreases the asymptotic time complexity of model building in hBOA by a factor of Theta(n(0.26)) to Theta(n(0.5)), where n is the problem size. On the other hand, sporadic model building also increases the number of evaluations until convergence;nonetheless, the evaluation slowdown is insignificant compared to the gains in the asymptotic complexity of model building.
To solve a wide range of different problems, the research in black-box optimization faces several important challenges. One of the most important challenges is the design of methods capable of automatic discovery and ...
详细信息
To solve a wide range of different problems, the research in black-box optimization faces several important challenges. One of the most important challenges is the design of methods capable of automatic discovery and exploitation of problem regularities to ensure efficient and reliable search for the optimum. This paper discusses the bayesian optimization algorithm (BOA), which uses bayesian networks to model promising solutions and sample new candidate solutions. Using bayesian networks in combination with population-based genetic and evolutionary search allows BOA to discover and exploit regularities in the form of a problem decomposition. The paper analyzes the applicability of the methods for learning bayesian networks in the context of genetic and evolutionary search and concludes that the combination of the two approaches yields robust, efficient, and accurate search. (C) 2002 Elsevier Science Inc. All rights reserved.
This paper applies the bayesian optimization algorithm with Tabu Search (Tabu-BOA) to electric equipment configuration problems in a power plant. Tabu-BOA is a hybrid evolutionary computation algorithm with competent ...
详细信息
This paper applies the bayesian optimization algorithm with Tabu Search (Tabu-BOA) to electric equipment configuration problems in a power plant. Tabu-BOA is a hybrid evolutionary computation algorithm with competent genetic algorithms and meta-heuristics. The configuration problems we consider have complex combinatorial properties, and therefore are hard to formulate and solve via conventional mathematical programming techniques. Using the proposed method, we have solved the following problems (in order of increasing complexity): (I) cost minimization of electric equipment configuration and the corresponding cabling, (2) plus choice of the power plant operation patterns, (3) plus parallel operation of multiple transformers, (4) plus change of the supply voltages (high voltage or low voltage) to the electric power load, and (5) with addition of another objective, that is, both minimization of the cost and maximization of the surplus power supply. (C) 2005 Wiley Periodicals, Inc.
This paper describes an evolutionary algorithm for optimization of continuous problems that combines advanced recombination techniques for discrete representations with advanced mutation techniques for continuous repr...
详细信息
This paper describes an evolutionary algorithm for optimization of continuous problems that combines advanced recombination techniques for discrete representations with advanced mutation techniques for continuous representations. Discretization is used to transform solutions between the discrete and continuous domains. The proposed algorithm combines the strengths of purely continuous and purely discrete approaches and eliminates some of their disadvantages. The paper tests the proposed algorithm with the recombination operator of the bayesian optimization algorithm, sigma-self-adaptive mutation, and three discretization methods. The empirical results on three problems suggest that the tested variant of the algorithm scales up well on all tested problems, indicating good scalability over a broad range of continuous problems. (C) 2003 Elsevier Inc. All rights reserved.
This article proposes a competent hierarchical optimization method called the hierarchical bayesian optimization algorithm (hBOA). hBOA extends the bayesian optimization algorithm (BOA) by incorporating three importan...
详细信息
The hybridization of genetic algorithms(GAs) and Tabu Search(TS) is one of the traditional problems in function optimization in the GA literature. However,most proposed methods so far have utilized GAs to explore glob...
详细信息
The hybridization of genetic algorithms(GAs) and Tabu Search(TS) is one of the traditional problems in function optimization in the GA literature. However,most proposed methods so far have utilized GAs to explore global candidates and TS to exploit local *** such methods,this paper discusses new algorithms to directly store individuals into multiple tabu lists during GA-iterations. The paper describes the basic idea,algorithms,experimental results, and their practical applications for social simulation and electric equipments design.
暂无评论