Neural network (NN) and evolutionary computation (EC) are two of the most popular and important techniques in computational intelligence, which can be combined together to solve the complex real world problems. This p...
详细信息
ISBN:
(纸本)9783319406633;9783319406626
Neural network (NN) and evolutionary computation (EC) are two of the most popular and important techniques in computational intelligence, which can be combined together to solve the complex real world problems. This paper represents a review of the researches that combined NN and EC. There are 3 main research focuses as follows. In the first research focus, EC algorithms have been widely used to optimize the structure and parameter of the NN, including weight, structure, learning rates, and others. In the second research focus, lots literatures have witnessed that EC based NNs are widespread in the applications such as classification, automatic control, prediction, and many other fields. These two kinds of researches into combining NN and EC are mainly focuses on using EC algorithms to optimize NN, to enhance the NN performance and the NN application ability. Our survey finds that particle swarm optimization is the most popular EC algorithm that the researchers choice to optimize NN in recent year, while genetic algorithm and differential evolution are also widely used. In the third research focus, there are also researches adopted NN as a tool to enhance the performance of EC algorithms. Although the literatures in this focus are not as many as the above two focuses, the existing results show that NN has great potential in enhancing EC algorithms. The survey shows that when NN and EC meet, combining them would result in an effective way to deal with the real world application. This interesting research topic has become more and more significant in the field of computer science and still has much room for development.
To study how Transfer Learning (TL) works and what are effective strategies for transfer learning, we propose to model the TL process using evolutionary computation. EC provides a clear model for a problem as searchin...
详细信息
ISBN:
(纸本)9781728183923
To study how Transfer Learning (TL) works and what are effective strategies for transfer learning, we propose to model the TL process using evolutionary computation. EC provides a clear model for a problem as searching through a set of potential solutions. We are able to more easily control and measure problem difficulty, problem similarity, and methods of information transfer and relate these to success. As a proof of concept, we will use a static source problem and three fixed target problems with simple known relationships (see Section III). We compare the effectiveness of several ways to transfer knowledge learned from solving one problem to solving the new problems in the context of the relationship between the problems. This we hope will demonstrate that using our EC model is a fruitful way to investigate TL. The results show there is an improvement for using some sampled methods representing the "learned knowledge" of the source problem S. Also, the results show that the diversity of the transferred population has some positive effect on finding the optimal solution depending on the relationship between source and target problems.
computational discovery of DNA motifs is one of the major challenges in bioinformatics, which helps in understanding the mechanism of gene regulation. It has been reported that computational approaches have good poten...
详细信息
ISBN:
(纸本)9781424481262
computational discovery of DNA motifs is one of the major challenges in bioinformatics, which helps in understanding the mechanism of gene regulation. It has been reported that computational approaches have good potential for problem solving in terms of cost and time saving. Based on our previous studies, this paper aims to develop an evolutionary computation scheme to provide an alternative approach for motif discovery. To work on the framework of our previously developed GAPK, a small sized collection of k-mers is extracted and utilized as "prior knowledge" in algorithm development. Our technical contributions in this paper mainly include a novel fitness function carrying information on conservation and rareness of DNA motifs, and a path to access GAPK-like solutions using seed concept and filtering techniques. The proposed algorithm in this paper has been evaluated by using eight benchmarked datasets, with comparisons to well-known tools such as MEME, MDScan, AlignACE and two GA-based techniques. Results show that our proposed method favorably outperforms other algorithms for these testing datasets.
In this paper, a method of automatic controller design for electronic control systems is described. In order to automate the design of an electronic controller, an evolutionary computation is applied. First, the frame...
详细信息
ISBN:
(纸本)9781424443475
In this paper, a method of automatic controller design for electronic control systems is described. In order to automate the design of an electronic controller, an evolutionary computation is applied. First, the framework for applying the genetic algorithm to the automation of controller design is described. In particular, the coding of a chromosome is shown in detail. Then, how to make a fitness function is represented, with an air conditioner as an example, and the controller of the air conditioner is developed automatically using our proposed framework. Finally, an evolutionary simulation is performed to confirm our framework.
Verification is increasingly becoming a bottleneck in the process of designing electronic circuits. While there exists several verification tools that assist in detecting occurrences of design errors, or bugs, there i...
详细信息
ISBN:
(纸本)9783662455234;9783662455227
Verification is increasingly becoming a bottleneck in the process of designing electronic circuits. While there exists several verification tools that assist in detecting occurrences of design errors, or bugs, there is a lack of solutions for accurately pin-pointing the root causes of these errors. Statistical bug localization has proven to be an approach that scales up to large designs and is widely utilized both in debugging hardware and software. However, the accuracy of localization is highly dependent on the quality of the stimuli. In this paper we formulate diagnostic test set generation as a task for an evolutionary algorithm, and propose dedicated fitness functions that closely correlate with the bug localization capabilities. We perform experiments on the register-transfer level design of the Plasma microprocessor coupling an evolutionary test-pattern generator and a simulator for fitness evaluation. As a result, the diagnostic resolution of the tests is significantly improved.
RSI is a commonly used indicator preferred by stock traders. However, even though it works well when the market is trendless, during bull or bear market conditions (when there is a clear trend) its performance degrade...
详细信息
RSI is a commonly used indicator preferred by stock traders. However, even though it works well when the market is trendless, during bull or bear market conditions (when there is a clear trend) its performance degrades. In this study, we developed a trading model using a modified RSI using trend-removed stock data. The model has several parameters including, the trend detection period, RSI buy-sell trigger levels and periods. These parameters are optimized using genetic algorithms;then the trading performance is compared against B&H and standard RSI indicator usage. 9 different ETFs are selected for evaluating trading performance. The results indicate there is a performance improvement both in profit and success rates using this new model. As future work, other indicators might be modelled in a similar fashion in order to see if it is possible to find one indicator that can work under any market condition. (C) 2014 The Authors. Published by Elsevier B.V.
The application of genetic and evolutionary computation to problems in medicine has increased rapidly over the past five years, but there are specific issues and challenges that distinguish it from other real-world ap...
详细信息
ISBN:
(纸本)9781450343237
The application of genetic and evolutionary computation to problems in medicine has increased rapidly over the past five years, but there are specific issues and challenges that distinguish it from other real-world applications. Obtaining reliable and coherent patient data, establishing the clinical need and demonstrating value in the results obtained are all aspects that require careful and detailed consideration.
作者:
Pei, YanUniv Aizu
Div Comp Sci Ikki Machi Fukushima 9658580 Japan
We consider algorithmic design, enhancement, and improvement of evolutionary computation (EC) as a mechanism design problem. All individuals or several groups of individuals can be considered as self-interested agents...
详细信息
ISBN:
(纸本)9781479986965
We consider algorithmic design, enhancement, and improvement of evolutionary computation (EC) as a mechanism design problem. All individuals or several groups of individuals can be considered as self-interested agents. The individuals in EC can manipulate the parameter settings and operations of an EC algorithm by satisfying their own preferences rather than by following a fixed algorithm rule. EC algorithm designers or EC self-adaptive methods should construct appropriate rules and mechanisms for all agents (individuals) to conduct their evolution behavior correctly in order to definitely achieve the desired and pre-set objective(s) definitively. We propose a formal framework on parameter setting, strategy selection, and algorithmic design of EC by considering the strategy equilibrium implementation of a mechanism design problem in the search process. We attempt to use Nash strategy equilibrium (NE) concept in an implementation of an algorithmic mechanism design problem, but our proposed framework is not limited to Nash strategy equilibrium. The evaluation results present the efficiency of the framework. Its primary principle can be implemented in any EC algorithm that needs to consider the strategy selection issue in its optimization process. The final objective of our work is to implement EC design as an algorithmic mechanism design problem and establish EC fundamental aspects based on this perspective.
In the paper the theoretical framework for cooperation and competition of coevolved population members working toward a common goal is presented. We use a formal model of evolutionary Turing Machine and its extensions...
详细信息
ISBN:
(纸本)9789898111050
In the paper the theoretical framework for cooperation and competition of coevolved population members working toward a common goal is presented. We use a formal model of evolutionary Turing Machine and its extensions to justify that in general evolutionary algorithms belong to the class of super-recursive algorithms. Parallel and Parallel Weighted evolutionary Turing Machine models have been proposed to capture properly cooperation and competition of the whole population expressed as an instance of multiobjective optimization.
Multimodal interfaces provide flexible, intuitive, and error-robust interaction with complex information systems. In this work we describe an innovative statistical approach for combining multimodal user input that is...
详细信息
ISBN:
(纸本)0780374029
Multimodal interfaces provide flexible, intuitive, and error-robust interaction with complex information systems. In this work we describe an innovative statistical approach for combining multimodal user input that is based on principles adopted from evolution theory. A population of individuals, each representing a solution to the integration problem, compete for an optimal interpretation of the user interactions. Specially designed genetic operators recombine various characteristics of these solutions. The fitness of a single individual, measuring the certainty and the confidence of an integration result, is calculated according to a weighted scheme including the various information resources and the current system context. Our integration algorithm works extremely robust. Moreover, it can easily be scaled up to additional input devices and various application domains.
暂无评论