• Refine Query
  • Source
  • Publication year
  • to
  • Language
  • 3
  • 3
  • Tagged with
  • 8
  • 8
  • 4
  • 3
  • 3
  • 2
  • 2
  • 2
  • 2
  • 2
  • 2
  • 2
  • 2
  • 2
  • 2
  • About
  • The Global ETD Search service is a free service for researchers to find electronic theses and dissertations. This service is provided by the Networked Digital Library of Theses and Dissertations.
    Our metadata is collected from universities around the world. If you manage a university/consortium/country archive and want to be added, details can be found on the NDLTD website.
1

Parameter estimation in tidally influenced numerical models:determination of an appropriate objective function

Tate, Jennifer N 09 August 2008 (has links)
The research detailed in this study focuses on the determination of an appropriate objective function to aid parameter estimation when simulating areas influenced by tidally varying flows. Three objective functions that are measures of how well the model results match field data at several locations and times were tested. A set of test cases is developed to represent tidally influenced systems and allow for the testing of the objective functions. These objective functions were tested by computing their values and comparing them for the various estimated parameters. Based on results of the first method of testing a further analysis was performed using PEST, an automatic parameter estimation tool. A weighted least squares of the velocity and water surface values with a weight function on the velocity term based on the shallow water equations is found to be a reasonable objective function at this point in the research.
2

A Study on Aggregation of Objective Functions in MaOPs Based on Evaluation Criteria

Furuhashi, Takeshi, Yoshikawa, Tomohiro, Otake, Shun January 2010 (has links)
Session ID: TH-E1-4 / SCIS & ISIS 2010, Joint 5th International Conference on Soft Computing and Intelligent Systems and 11th International Symposium on Advanced Intelligent Systems. December 8-12, 2010, Okayama Convention Center, Okayama, Japan
3

Improving Neural Network Classification Training

Rimer, Michael Edwin 05 September 2007 (has links) (PDF)
The following work presents a new set of general methods for improving neural network accuracy on classification tasks, grouped under the label of classification-based methods. The central theme of these approaches is to provide problem representations and error functions that more directly improve classification accuracy than conventional learning and error functions. The CB1 algorithm attempts to maximize classification accuracy by selectively backpropagating error only on misclassified training patterns. CB2 incorporates a sliding error threshold to the CB1 algorithm, interpolating between the behavior of CB1 and standard error backpropagation as training progresses in order to avoid prematurely saturated network weights. CB3 learns a confidence threshold for each combination of training pattern and output class. This models an error function based on the performance of the network as it trains in order to avoid local overfit and premature weight saturation. PL1 is a point-wise local binning algorithm used to calibrate a learning model to output more accurate posterior probabilities. This algorithm is used to improve the reliability of classification-based networks while retaining their higher degree of classification accuracy. These approaches are demonstrated to be robust to a variety of learning parameter settings and have better classification accuracy than standard approaches on a variety of applications, such as OCR and speech recognition.
4

Structural Condition Assessment of Steel Stringer Highway Bridges

Wang, Xiaoyi 13 July 2005 (has links)
No description available.
5

Trends and Observations from Steel Stringer Bridge Model Calibrations

Barber, Matthew Gabriel January 2008 (has links)
No description available.
6

Reavaliação rápida em problemas de otimização quadrática binária

Anacleto, Eduardo Alves de Jesus January 2018 (has links)
Orientador: Prof. Dr. Cláudio Nogueira de Meneses / Dissertação (mestrado) - Universidade Federal do ABC, Programa de Pós-Graduação em Ciência da Computação, 2018. / Diversos problemas da area de otimização combinatoria podem ser convertidos, em tempo polinomial, para o problema de Programação Quadratica Binaria Irrestrita (UBQP). Neste problema, desejamos encontrar um vetor solução binario x, de dimensão n, tal que a função objetivo f(x) = x|Qx tenha valor mínimo, onde Q é uma matriz com coeficientes racionais. Em termos de complexidade computacional, o problema UBQP pertence a classe NP-difícil. A importancia deste problema, tanto pratica quanto teorica, tem motivado muitos pesquisadores a dedicarem uma quantidade razoavel de tempo tentando projetar tecnicas de resolução exatas e heuristicas para este problema. Durante o processo de resolução do problema UBQP, estas tecnicas necessitam reavaliar muitas vezes o valor da função objetivo. Dependendo da maneira como esta reavaliação é realizada, pode ser preciso executar um numero relativamente grande de operações elementares (atribuições, adições, subtrações e comparações). Isto pode consumir muito tempo de processamento quando n é grande. Nesta pesquisa, propomos formulas que requerem poucas operações para efetuar a reavaliação. Na literatura do problema UBQP, formulas de reavaliação são aplicadas, normalmente, quando há ate duas alterações nos componentes do vetor solução. As formulas que deduzimos podem ser usadas para efetuar qualquer quantidade de alterações. Analisamos uma das nossas formulas de maneira teorica e deduzimos funções que podem ser adotadas para indicar o melhor momento para aplicar essa formula. Ademais, projetamos algoritmos com estas formulas de reavaliação e verificamos a praticidade destes algoritmos conduzindo experimentos computacionais usando implementações de heurísticas de busca local e Variable Neighborhood Search. Nesses experimentos comparamos o desempenho dessas implementações ao resolver instancias da literatura para o problema UBQP. Os resultados experimentais evidenciaram que as formulas de reavaliação, propostas, podem propiciar reduções relativamente grandes nos tempos de processamento, mesmo quando o numero de diferenças entre soluções é moderadamente grande. / Several combinatorial optimization problems can be reformulated, in polynomial time, to the Unconstrained Binary Quadratic Programming (UBQP) problem. In this problem, we are interested in finding an n-dimensional binary solution vector, x, that minimizes the objective function f(x) = x|Qx, where Q is a matrix with rational coecients. In terms of computational complexity, the UBQP problem belongs to the NP-hard class. The practical and theoretical importance of this problem has motivated many researchers to dedicate a reasonable amount of time developing exact and heuristic solution techniques to solve this problem. During the resolution process of the UBQP problem, these techniques need to evaluate many times the objective function value. Depending on how it is made, it may be necessary to execute a relatively large number of elementary operations, such as assignments, additions, subtractions and comparisons. For n large, this may be time consuming. In this research, we propose formulas to perform the reevaluation requiring lesser operations than the simple evaluation of the objective function. In the literature of the UBQP problem, it is common to use reevaluation formulas only when there are at most two- ip moves that simultaneously change the values of two components. The formulas we have deduced can be used to evaluate any number of ip moves. We analyzed one of our reevaluation formulas and deduced functions that can be used to suggest the best moment to apply this formula. In addition, we designed algorithms with these reevaluation formulas and verified the practicality of these algorithms by conducting computational experiments using implementations of local search and Variable Neighborhood Search heuristics. In these experiments, we compared the performance of these implementations by solving benchmark instances for the UBQP problem. The experimental results showed that the reevaluation formulas we created can provide relatively large reductions in processing times, even when the number of ip moves is moderately large.
7

Aplicação interativa em processos de otimização por método das estratégias de evolução / Interactive application in optimization processes by evolution strategies method

Jesus, Luiz Henrique Reis de 15 May 2017 (has links)
Submitted by Luciana Ferreira (lucgeral@gmail.com) on 2017-06-14T11:09:20Z No. of bitstreams: 2 license_rdf: 0 bytes, checksum: d41d8cd98f00b204e9800998ecf8427e (MD5) Dissertação - Luiz Henrique Reis de Jesus - 2017.pdf: 3311759 bytes, checksum: 20e4044376c2666f102131892f7b3fc2 (MD5) / Approved for entry into archive by Luciana Ferreira (lucgeral@gmail.com) on 2017-06-14T11:09:35Z (GMT) No. of bitstreams: 2 license_rdf: 0 bytes, checksum: d41d8cd98f00b204e9800998ecf8427e (MD5) Dissertação - Luiz Henrique Reis de Jesus - 2017.pdf: 3311759 bytes, checksum: 20e4044376c2666f102131892f7b3fc2 (MD5) / Made available in DSpace on 2017-06-14T11:09:35Z (GMT). No. of bitstreams: 2 license_rdf: 0 bytes, checksum: d41d8cd98f00b204e9800998ecf8427e (MD5) Dissertação - Luiz Henrique Reis de Jesus - 2017.pdf: 3311759 bytes, checksum: 20e4044376c2666f102131892f7b3fc2 (MD5) Previous issue date: 2017-05-15 / Conselho Nacional de Pesquisa e Desenvolvimento Científico e Tecnológico - CNPq / This dissertation of master degree describes an approach of the interactive optimization process associate to the Evolution Strategies method for the evaluation of the loudspeaker optimization project, with the objective to present the advantages achieved after the user interventions throughout the optimization process. Its development is based on the methodology of the Evolution Strategies characterized by the concept of natural selection, which uses combination and mutation methods to generate new individuals. However, for greater efficiency in the responses of the objective function, as well as the reduction in its convergence time, the optimization process requires interventions in stabilization of local minima and maxima. In the interventions made available to the user, will be considered manipulations in the operators of mutation and combination, mutation of the parameters of self-adaptation, as well as the change of objective and the variation of their respective restriction. As a differential, an interface was developed to make feasible the user interventions applied to the optimization process, as well as the monitoring of the entire process. In this work, also evaluated optimization test functions with the objective of validating the proposed methodology. / Esta dissertação de mestrado descreve uma abordagem do processo de otimização interativa associado ao método das Estratégias de Evolução para a avaliação do projeto de otimização do alto-falante, com o objetivo de apresentar as vantagens alcançadas após as intervenções do usuário ao longo do processo de otimização. Seu desenvolvimento é baseado na metodologia das Estratégias de Evolução caracterizada pelo conceito de seleção natural, o qual utiliza de métodos de combinação e mutação para a geração de novos indivíduos. No entanto, para maior eficiência nas respostas a função objetivo, bem como a redução em seu tempo de convergência, o processo de otimização necessita de intervenções em estabilizações de mínimos e máximos locais. Nas intervenções disponibilizadas ao usuário, serão consideradas manipulações nos operadores de mutação e combinação, mutação dos parâmetros de auto-adaptação, bem como a mudança de objetivo e a variação de sua respectiva restrição. Como diferencial, foi desenvolvida uma interface para viabilizar as intervenções do usuário aplicadas ao processo de otimização, bem como o acompanhamento de todo o processo. Neste trabalho, também foram avaliadas funções de teste de otimização com o objetivo de validar a metodologia proposta.
8

[en] INVERSE OPTIMIZATION VIA ONLINE LEARNING / [pt] OTIMIZAÇÃO INVERSA VIA ONLINE LEARNING

LUISA SILVEIRA ROSA 02 April 2020 (has links)
[pt] Demonstramos como aprender a função objetivo e as restrições de problemas de otimização enquanto observamos sua solução ótima no decorrer de múltiplas rodadas. Nossa abordagem é baseada em técnicas de Online Learning e funciona para funções objetivo lineares sob conjuntos viáveis arbitrários generalizando trabalhos anteriores. Os dois algoritmos, um para aprender a função objetivo e o outro par aprender as restrições, convergem a uma taxa de O (1 sobre raiz de T) que nos permitem produzir soluções tão boas quanto as ótimas em poucas observações. Finalmente, mostramos a eficácia e possíveis aplicações de nossos métodos em um amplo estudo computacional. / [en] We demonstrate how to learn the objective function and constraints of optimization problems while observing its optimal solution over multiple rounds. Our approach is based on Online Learning techniques and works for linear objective functions under arbitrary feasible sets by generalizing previous work. The two algorithms, one to learn objective function and other to learn constraints, converge at a rate of O (1 on t root) that allow us to produce solutions as good as the optimal in a few observations. Finally, we show the efficacy and possible applications of our methods in a significant computational study.

Page generated in 0.1077 seconds