oalib
Search Results: 1 - 10 of 100 matches for " "
All listed articles are free for downloading (OA Articles)
Page 1 /100
Display every page Item
Optimization of Evolutionary Neural Networks Using Hybrid Learning Algorithms  [PDF]
Ajith Abraham
Computer Science , 2004,
Abstract: Evolutionary artificial neural networks (EANNs) refer to a special class of artificial neural networks (ANNs) in which evolution is another fundamental form of adaptation in addition to learning. Evolutionary algorithms are used to adapt the connection weights, network architecture and learning algorithms according to the problem environment. Even though evolutionary algorithms are well known as efficient global search algorithms, very often they miss the best local solutions in the complex solution space. In this paper, we propose a hybrid meta-heuristic learning approach combining evolutionary learning and local search methods (using 1st and 2nd order error information) to improve the learning and faster convergence obtained using a direct evolutionary approach. The proposed technique is tested on three different chaotic time series and the test results are compared with some popular neuro-fuzzy systems and a recently developed cutting angle method of global optimization. Empirical results reveal that the proposed technique is efficient in spite of the computational complexity.
Evolutionary Algorithm for Optimal Connection Weights in Artificial Neural Networks  [PDF]
G.V.R. Sagar, S. Venkata Chalam, Manoj Kumar Singh
International Journal of Engineering , 2011,
Abstract: A neural network may be considered as an adaptive system that progressively self-organizes inorder to approximate the solution, making the problem solver free from the need to accuratelyand unambiguously specify the steps towards the solution. Moreover, Evolutionary computationcan be integrated with artificial Neural Network to increase the performance at various levels; inresult such neural network is called Evolutionary ANN. In this paper very important issue of neuralnetwork namely adjustment of connection weights for learning presented by Genetic algorithmover feed forward architecture. To see the performance of developed solution comparison hasgiven with respect to well established method of learning called gradient decent method. Abenchmark problem of classification, XOR, has taken to justify the experiment. Presented methodis not only having very probability to achieve the global minima but also having very fastconvergence.
EvoNF: A Framework for Optimization of Fuzzy Inference Systems Using Neural Network Learning and Evolutionary Computation  [PDF]
Ajith Abraham
Computer Science , 2004,
Abstract: Several adaptation techniques have been investigated to optimize fuzzy inference systems. Neural network learning algorithms have been used to determine the parameters of fuzzy inference system. Such models are often called as integrated neuro-fuzzy models. In an integrated neuro-fuzzy model there is no guarantee that the neural network learning algorithm converges and the tuning of fuzzy inference system will be successful. Success of evolutionary search procedures for optimization of fuzzy inference system is well proven and established in many application areas. In this paper, we will explore how the optimization of fuzzy inference systems could be further improved using a meta-heuristic approach combining neural network learning and evolutionary computation. The proposed technique could be considered as a methodology to integrate neural networks, fuzzy inference systems and evolutionary search procedures. We present the theoretical frameworks and some experimental results to demonstrate the efficiency of the proposed technique.
On the Relationships between Generative Encodings, Regularity, and Learning Abilities when Evolving Plastic Artificial Neural Networks  [PDF]
Paul Tonelli, Jean-Baptiste Mouret
PLOS ONE , 2013, DOI: 10.1371/journal.pone.0079138
Abstract: A major goal of bio-inspired artificial intelligence is to design artificial neural networks with abilities that resemble those of animal nervous systems. It is commonly believed that two keys for evolving nature-like artificial neural networks are (1) the developmental process that links genes to nervous systems, which enables the evolution of large, regular neural networks, and (2) synaptic plasticity, which allows neural networks to change during their lifetime. So far, these two topics have been mainly studied separately. The present paper shows that they are actually deeply connected. Using a simple operant conditioning task and a classic evolutionary algorithm, we compare three ways to encode plastic neural networks: a direct encoding, a developmental encoding inspired by computational neuroscience models, and a developmental encoding inspired by morphogen gradients (similar to HyperNEAT). Our results suggest that using a developmental encoding could improve the learning abilities of evolved, plastic neural networks. Complementary experiments reveal that this result is likely the consequence of the bias of developmental encodings towards regular structures: (1) in our experimental setup, encodings that tend to produce more regular networks yield networks with better general learning abilities; (2) whatever the encoding is, networks that are the more regular are statistically those that have the best learning abilities.
Learning limits of an artificial neural network
Vega, J.J;Reynoso, R;Carrillo Calvet, H;
Revista mexicana de física , 2008,
Abstract: technological advances in hardware as well as new computational paradigms give us the opportunity to apply digital techniques to pulse shape analysis (psa), requiring powerful resources. in this paper, we present a psa application based on artificial neural networks (anns). these adaptive systems offer several advantages for these tasks; nevertheless it is necessary to face the particular problems linked to them as: the selection of the learning rule and the ann architecture, the sizes of the training and validation data sets, overtraining, the effect of noise on the pattern identification ability, etc. we will present evidences of the effect on the performance of a back-propagation ann as a pattern identifier of both: the size of the noise that the bragg curve spectrometer signal present and of overtraining. in fact, these two effects are related.
Learning limits of an artificial neural network
J.J. Vega,R. Reynoso,H. Carrillo Calvet
Revista mexicana de física , 2008,
Abstract: Technological advances in hardware as well as new computational paradigms give us the opportunity to apply digital techniques to Pulse Shape Analysis (PSA), requiring powerful resources. In this paper, we present a PSA application based on Artificial Neural Networks (ANNs). These adaptive systems offer several advantages for these tasks; nevertheless it is necessary to face the particular problems linked to them as: the selection of the learning rule and the ANN architecture, the sizes of the training and validation data sets, overtraining, the effect of noise on the pattern identification ability, etc. We will present evidences of the effect on the performance of a back-propagation ANN as a pattern identifier of both: the size of the noise that the Bragg curve spectrometer signal present and of overtraining. In fact, these two effects are related.
LEARNING PHENOMENA IN MANUFACTURING AND ARTIFICIAL NEURAL NETWORK  [cached]
Miroslav Car
Journal of Information and Organizational Sciences , 2007,
Abstract: In the industrially advanced countries, that are different from our ex and present countries, to learning phenomena has been dedicated a significant attention for the last 60 years. One of more basic reasons is multiple purposes of results. Until now, there have been applied various approaches, methods and procedures for empirical data approximation, and in this articles some possibilities of artificial neural network application are researched.
Evolutionary Artificial Neural Network Based on Chemical Reaction Optimization  [PDF]
James J. Q. Yu,Albert Y. S. Lam,Victor O. K. Li
Computer Science , 2015, DOI: 10.1109/CEC.2011.5949872
Abstract: Evolutionary algorithms (EAs) are very popular tools to design and evolve artificial neural networks (ANNs), especially to train them. These methods have advantages over the conventional backpropagation (BP) method because of their low computational requirement when searching in a large solution space. In this paper, we employ Chemical Reaction Optimization (CRO), a newly developed global optimization method, to replace BP in training neural networks. CRO is a population-based metaheuristics mimicking the transition of molecules and their interactions in a chemical reaction. Simulation results show that CRO outperforms many EA strategies commonly used to train neural networks.
Artificial neural networks and evolutionary algorithms in engineering design  [PDF]
T. Velsker,M. Eerme,J. Majak,M. Pohlak
Journal of Achievements in Materials and Manufacturing Engineering , 2011,
Abstract: Purpose: Purpose of this paper is investigation of optimization strategies eligible for solving complex engineering design problems. An aim is to develop numerical algorithms for solving optimal design problems which may contain real and integer variables, a number of local extremes, linear- and non-linear constraints and multiple optimality criteria.Design/methodology/approach: The methodology proposed for solving optimal design problems is based on integrated use of meta-modeling techniques and global optimization algorithms. Design of the complex and safety critical products is validated experimentally.Findings: Hierarchically decomposed multistage optimization strategy for solving complex engineering design problems is developed. A number of different non-gradient methods and meta-modeling techniques has been evaluated and compared for certain class of engineering design problems. The developed optimization algorithms allows to predict the performance of the product (structure) for different design and configurations parameters as well as loading conditions.Research limitations/implications: The results obtained can be applied for solving certain class of engineering design problems. The nano- and microstructure design of materials is not considered in current approach.Practical implications: The methodology proposed is employed successfully for solving a number of practical problems arising from Estonian industry: design of car frontal protection system, double-curved surface forming process modeling, fixings for frameless glazed structures, optimal design of composite bathtub (large composite plastics), etc.Originality/value: Developed numerical algorithms can be utilised for solving a wide class of complex optimization problems.
E-Learning Optimization Using Supervised Artificial Neural-Network  [PDF]
Mohamed Sayed, Faris Baker
Journal of Software Engineering and Applications (JSEA) , 2015, DOI: 10.4236/jsea.2015.81004
Abstract: Improving learning outcome has always been an important motivating factor in educational inquiry. In a blended learning environment where e-learning and traditional face to face class tutoring are combined, there are opportunities to explore the role of technology in improving student’s grades. A student’s performance is impacted by many factors such as engagement, self-regulation, peer interaction, tutor’s experience and tutors’ time involvement with students. Furthermore, e-course design factors such as providing personalized learning are an urgent requirement for improved learning process. In this paper, an artificial neural network model is introduced as a type of supervised learning, meaning that the network is provided with example input parameters of learning and the desired optimized and correct output for that input. We also describe, by utilizing e-learning interactions and social analytics how to use artificial neural network to produce a converging mathematical model. Then students’ performance can be efficiently predicted and so the danger of failing in an enrolled e-course should be reduced.
Page 1 /100
Display every page Item


Home
Copyright © 2008-2017 Open Access Library. All rights reserved.