Many real-world optimization problems involve two different subsets of variables: decision variables, and those variables which are not present in the cost function but constrain the solutions, and thus, must be consi...
详细信息
Many real-world optimization problems involve two different subsets of variables: decision variables, and those variables which are not present in the cost function but constrain the solutions, and thus, must be considered during optimization. Thus, dependencies between and within both subsets of variables must be considered. In this paper, an estimation of distribution algorithm (EDA) is implemented to solve this type of complex optimization problems. A Gaussian Bayesian network is used to build an abstraction model of the search space in each iteration to identify patterns among the variables. As the algorithm is initialized from data, we introduce a new hyper-parameter to control the influence of the initial data in the decisions made during the EDA execution. The results show that our algorithm improves the cost function more than the expert knowledge does.
Regression learning methods in real world applications often require cost minimization instead of the reduction of various metrics of prediction errors. Currently in the literature, there is a lack of white box soluti...
详细信息
Regression learning methods in real world applications often require cost minimization instead of the reduction of various metrics of prediction errors. Currently in the literature, there is a lack of white box solutions that can deal with forecasting problems where under-prediction and over-prediction errors have different consequences. To fill this gap, we introduced the Cost-sensitive Global Model Tree (CGMT), which applies a fitness function that minimizes an average misprediction cost. Proposed specialized genetic operators improve searching for optimal tree structure and cost-sensitive linear regression models in the leaves. Experimental validation is performed on loan charge-off data. It is known to be a difficult forecasting problem for banks due to the asymmetric cost structure. Obtained results show that specialized evolutionary algorithm applied to model tree induction finds significantly more accurate predictions than tested competitors. Decisions generated by the CGMT are simple, easy to interpret, and can be applied directly. (C) 2015 Elsevier B.V. All rights reserved.
evolutionary computing is an exciting sub-field of soft computing. Many evolutionary algorithm based on the Darwinian principles of natural selection are developed under the umbrella of EC in the last two decades. EAs...
详细信息
evolutionary computing is an exciting sub-field of soft computing. Many evolutionary algorithm based on the Darwinian principles of natural selection are developed under the umbrella of EC in the last two decades. EAs provide a set of optimal solutions in single simulation unlike traditional optimization techniques for dealing with large-scale global optimization and search problems. Teaching Learning based Optimization (TLBO) is one of the most recently developed EA. TLBO employs a group of learners or a class of learners to perform global optimization search process. The framework of the TLBO consists of two phases, including the Teacher Phase and Learner Phase. The Teacher Phase' means learning from the teachers and the Learner Phase means learning through interaction among learners. In this paper, we have developed a hybrid TLBO (HTLBO) with aim at to further improve the exploration and exploitation abilities of the baseline TLBO algorithm. The performance of the proposed HTLBO algorithm examined upon using recently designed benchmark functions for the special session of the CEC2017 problems. The experimental results of the proposed algorithm are better than some well-known evolutionary algorithms in terms of proximity and diversity. (C) 2021 THE AUTHORS. Published by Elsevier BV on behalf of Faculty of Engineering, Alexandria University.
This work focuses on multi-objective evolutionary optimization by approximation function. It uses the new general concept of evolution control to on-line enriching the database of correct solutions, which are the basi...
详细信息
The proceedings contain 124 papers from the 2005 IEEE Congress on evolutionary Computation, IEEE CEC 2005. Proceedings. The topics discussed include: adaptive cluster covering and evolutionary approach: comparison, di...
详细信息
ISBN:
(纸本)0780393635
The proceedings contain 124 papers from the 2005 IEEE Congress on evolutionary Computation, IEEE CEC 2005. Proceedings. The topics discussed include: adaptive cluster covering and evolutionary approach: comparison, differences and similarities;evolving improved incremental learning schemes for neural network systems;a hybrid ant colony optimization approach (hACO) for constructing load-balanced clusters;building anticipations in an accuracy-based learning classifier system by use of an artificial neural network;and applying genetic programming to learn spatial differences between textures using a translation invariant representation.
A number of evolutionary Computations (ECs) have been developed for solving Multimodal Function Optimization Problems (MFOPs) [1]. Some of the well-known ones are: Fitness Sharing [2], Sequential Niching [3], Simple S...
详细信息
ISBN:
(纸本)1889335185
A number of evolutionary Computations (ECs) have been developed for solving Multimodal Function Optimization Problems (MFOPs) [1]. Some of the well-known ones are: Fitness Sharing [2], Sequential Niching [3], Simple Subpopulation Schemes [4] and Co-evolutionary Shared Niching [5]. These ECs have shown the capability of solving MFOPs, but have introduced one or more parameters that cannot be easily set without prior knowledge of the fitness landscape. Moreover, a priori knowledge of a particular MFOP may not always be readily available. In this work, we describe a set of parallel and distributed ECs that are capable of locating all the peaks in a MFOP without using parameters that require specific topological information. This paper also provides a performance comparison between three approaches to solving MFOPs: Fitness Sharing, parallel EC and distributed EC.
A local navigation algorithm for mobile robots is proposed, based on the new extended virtual force field (EVFF) concept, neural network-based fusion for the three primitive behaviors generated by the EVFF, and the ev...
详细信息
A multi-objective evolutionary algorithm is used to deter- mine the membership function distribution within the outer loop control system of a non-linear missile autopilot using lateral acceleration con- trol. This pr...
详细信息
The proceedings contain 66 papers from the Applications on evolutionary Computing - EvoWorkshops 2005: EvoBIO, EvoCOMNET, EvoHOT, EvoIASP, EvoMUSART, and EvoSTOC, Proceedings. The topics discussed include: a fuzzy vit...
详细信息
The proceedings contain 66 papers from the Applications on evolutionary Computing - EvoWorkshops 2005: EvoBIO, EvoCOMNET, EvoHOT, EvoIASP, EvoMUSART, and EvoSTOC, Proceedings. The topics discussed include: a fuzzy viterbi algorithm for improved sequence alignment and searching of proteins;order preserving clustering over multiple time course experiments;neural networks and temporal gene expression data;bayesian learning with local support vector machines for cancer classification with gene expression data;syntactic approach to predict membrane spanning regions of transmembrane proteins;and an evolutionary approach for motif discovery and transmembrane protein classification.
-This paper provides a computational methodology based on monarch butterfly optimization (MBO) to find a solution to the problem of cost-based unit commitment (CBUC). The binary variables of unit commitment problems a...
详细信息
-This paper provides a computational methodology based on monarch butterfly optimization (MBO) to find a solution to the problem of cost-based unit commitment (CBUC). The binary variables of unit commitment problems are handled by modifying the continuous-time nature of the monarch butterfly algorithm. Thermal unit generation, uptime, downtime, ramp rate limits as well as system reserve are considered in the test systems. The computational approach has many parts that not only minimize the cost function but also handle the mixed constraints of the commitment problem. The effect of thermal turbine valve-point loading is also taken into consideration. The computational technique has been used to solve a ten-unit original system and five scaled-up adaptations obtained from this base system. The results obtained are in agreement with the recent results available in the literature. Comparative analysis shows the effectiveness of the proposed MBO-based solution methodology in terms of operating costs and execution time in relation to other techniques.
暂无评论