We describe first results on a traffic-lights controller based on neural networks optimized by an evolutionary algorithm. Among the inputs of the neural network are outputs of other popular control algorithms, thus th...
详细信息
ISBN:
(纸本)9781538674499
We describe first results on a traffic-lights controller based on neural networks optimized by an evolutionary algorithm. Among the inputs of the neural network are outputs of other popular control algorithms, thus the evolved controller can be considered and ensemble controller. In a series of experiments, we show that evolution is capable of creating controllers that provide promising performance better than any of a number of baselines.
In this paper, the problem of obtaining optimal routes on tridimensional environments is discussed. This scenario is called as Traveler Salesman Problem (TSP 3D-variation). As is widely known, TSP has NP-complexity so...
详细信息
In this paper, the problem of obtaining optimal routes on tridimensional environments is discussed. This scenario is called as Traveler Salesman Problem (TSP 3D-variation). As is widely known, TSP has NP-complexity so is necessary to apply techniques to solve it approximately (no exacts solutions available). The purpose of this research is to present a genetic algorithm to solve 3D-TSP variation. These kind of evolutionary algorithms are ideal for solving complex problems where necessary rearrangements and route optimization. In case of genetic algorithms, optimal solutions appear faster depending on the quality of initial population, so theory recommends using metaheuristics for generating this population. In this study, it has used a metaheuristic GRASP algorithm to generate the initial population and, over it, apply the genetic operators proposed for optimizing individuals obtained. The results have optimal routes of movement and displacement and are directly applicable in the storage industry.
Comparative study of two evolutionary algorithms namely backtracking search algorithm (BSA) and firefly algorithm (FA) has been described in this paper for synthesis of cosecant squared radiation pattern in linear ant...
详细信息
The aging of the voice, known as presbyphonia, is a natural process that can cause great change in vocal quality of the individual. This is a relevant problem to those people who use their voices professionally, and i...
详细信息
The aging of the voice, known as presbyphonia, is a natural process that can cause great change in vocal quality of the individual. This is a relevant problem to those people who use their voices professionally, and its early identification can help determine a suitable treatment to avoid its progress or even to eliminate the problem. This work focuses on the development of a new model for the identification of aging voices (independently of their chronological age), using as input attributes parameters extracted from the voice and glottal signals. The proposed model, named Quantum binary-real evolving Spiking Neural Network (QbrSNN), is based on spiking neural networks (SNNs), with an unsupervised training algorithm, and a Quantum-Inspired evolutionary Algorithm that automatically determines the most relevant attributes and the optimal parameters that configure the SNN. The QbrSNN model was evaluated in a database composed of 120 records, containing samples from three groups of speakers. The results obtained indicate that the proposed model provides better accuracy than other approaches, with fewer input attributes.
The evaluation of the comprehensive quality of college students is a key problem in the management of college student affairs. In this paper, we present an improved memetic differential evolution algorithm to get the ...
详细信息
This paper presents a research on the calculation of short- to mid-range ship routes that are based on density maps derived by previous historical locations of liner or merchant ships. Two main approaches are presente...
详细信息
This paper presents a research on the calculation of short- to mid-range ship routes that are based on density maps derived by previous historical locations of liner or merchant ships. Two main approaches are presented. In the first one, the route finding problem is formulated as an evolutionary-optimization problem. In the second one, a modified A* algorithm is presented, which is able to handle density data and smoothing requirements. Both methods are able to calculate accurate and smooth ship routes that comply with exiting density data of common sea paths. A combination of both methods is also presented for deriving smooth ship routes that comply with density data without the need for post processing. Several examples are presented and discussed to illustrate the effectiveness and the performance of all proposed methods.
In this paper, we consider a capacitated multiple allocation hub location problem derived from a practical application in network design of German wagonload traffic. Due to the difficulty to solve even small data sets...
详细信息
In this paper, we consider a capacitated multiple allocation hub location problem derived from a practical application in network design of German wagonload traffic. Due to the difficulty to solve even small data sets to optimality, we present two matheuristics: a local search matheuristic and an extension of an evolutionary algorithm matheuristic. Computational results are presented to demonstrate and compare the efficiency of both approaches for real-sized instances.
An algorithm to perform mate selection in aquaculture breeding using a computational optimization procedure called "differential evolution" (DE) was applied under optimum contribution selection and mate sele...
详细信息
An algorithm to perform mate selection in aquaculture breeding using a computational optimization procedure called "differential evolution" (DE) was applied under optimum contribution selection and mate selection scenarios, to assess its efficiency in maximizing the genetic merit while controlling inbreeding. Real aquaculture data sets with 8,782 Nile tilapias from five generations and 79,144 coho salmon from eight generations were used to optimize objective functions accounting for coancestry of parents and expected genetic merit and inbreeding of the future progeny. The mate selection results were compared with those from the realized scenario (real mates), truncation selection and optimum contribution selection method. Mate selection allowed reducing inbreeding up to 73% for Nile tilapia, compared with truncation selection, and up to 20% for coho salmon, compared with realized scenario. There was evidence that mate selection outperformed optimum contribution selection followed by minimum inbreeding mating in controlling inbreeding under the same expected genetic gain. The developed algorithm was computationally efficient in maximizing the objective functions and flexible for practical application in aquaculture breeding.
Association rule mining is one of the most important data mining tasks. It corresponds to the determination of rules that associate items to other items in a data set, where the items are attributes in transactional d...
详细信息
ISBN:
(纸本)9783642353796;9783642353802
Association rule mining is one of the most important data mining tasks. It corresponds to the determination of rules that associate items to other items in a data set, where the items are attributes in transactional databases. Although evolutionary algorithms have been used in this task for some time, there are few applications of immune algorithms to such problem. This paper presents one typical genetic algorithm plus two clonal selection algorithms applied to association rule mining under the perspective of several measures of interest.
Striking an effective balance between exploration and exploitation (E&E) is still one of the major concerns when using evolutionary algorithms (EAs) in dynamic environments. In this work, a new scheme for adaptive...
详细信息
ISBN:
(纸本)9781450349390
Striking an effective balance between exploration and exploitation (E&E) is still one of the major concerns when using evolutionary algorithms (EAs) in dynamic environments. In this work, a new scheme for adaptively balancing E&E in EAs is proposed. Based on the results of a statistical Pre-Post analysis of the population, the next search mode can be decided (i.e., exploration or exploitation). The experimental results showed that our proposal excels versus several competing approaches from the state of the art.
暂无评论