Дисертації з теми "Recurrent Problems"
Оформте джерело за APA, MLA, Chicago, Harvard та іншими стилями
Ознайомтеся з топ-40 дисертацій для дослідження на тему "Recurrent Problems".
Біля кожної праці в переліку літератури доступна кнопка «Додати до бібліографії». Скористайтеся нею – і ми автоматично оформимо бібліографічне посилання на обрану працю в потрібному вам стилі цитування: APA, MLA, «Гарвард», «Чикаго», «Ванкувер» тощо.
Також ви можете завантажити повний текст наукової публікації у форматі «.pdf» та прочитати онлайн анотацію до роботи, якщо відповідні параметри наявні в метаданих.
Переглядайте дисертації для різних дисциплін та оформлюйте правильно вашу бібліографію.
Meister, Daniel. "The complexity of membership problems for finite recurrent systems and minimal triangulations." Doctoral thesis, [S.l.] : [s.n.], 2006. http://deposit.ddb.de/cgi-bin/dokserv?idn=981745369.
Повний текст джерелаSingh, Jayant. "Optimization Problems Arising in Stability Analysis of Discrete Time Recurrent Neural Networks." Diss., North Dakota State University, 2016. http://hdl.handle.net/10365/25537.
Повний текст джерелаBell, Tannisha D. "An Examination of Race and Recurrent Substance Problems in the United States." Thesis, Virginia Tech, 2000. http://hdl.handle.net/10919/9582.
Повний текст джерелаMaster of Science
Nakayama, Masataka. "The problems of serial order in language:Clustering, context discrimination, temporal distance, and edges." Kyoto University, 2015. http://hdl.handle.net/2433/200480.
Повний текст джерелаJeffrey, James Howard. "Recurrent problems in the construction process : an action research study of the design and construction quality of hospital en-suites." Thesis, University of Salford, 2010. http://usir.salford.ac.uk/26736/.
Повний текст джерелаCAROTA, MASSIMO. "Neural network approach to problems of static/dynamic classification." Doctoral thesis, Università degli Studi di Roma "Tor Vergata", 2008. http://hdl.handle.net/2108/580.
Повний текст джерелаMossina, Luca. "Applications d'apprentissage automatique à la résolution de problèmes récurrents en optimisation combinatoire." Electronic Thesis or Diss., Toulouse, ISAE, 2020. http://www.theses.fr/2020ESAE0043.
Повний текст джерелаThe interest is on those decision problems for which an optimal or quasi-optimal solution is sought, and for which it is necessary to solve successive instances (recurrent problems) that are variations of a common original problem.The structure of such problems is analysed to identify the characteristics that can be exploited and transferred from one resolution to another, to incrementally improve the quality of the optimization process. The research is characterized by the interaction between a process of statistical learning (from optimization data) and a process of optimization. The information extracted from past resolutions is generalized to the current problem and integrated into the optimization algorithm to make its execution more resource-efficient.In particular, this thesis presents three contributions.The first, introduces a method that generates a simpler sub-problem to an instance of a recurrent problem, using multi-label classification. A subset of decision variables is selected and set to a reference value. The solution to the remaining sub-problem, while not guaranteed to be optimal for the original problem, can be obtained faster.The second employs Supervised Learning, classification and regression, to predict an additional constraint to a reference recurrent problem modelled via Mathematical Programming. When a new instance is solved, the model predicts how much of the solution to the reference problem is still applicable, allowing for a more rapid resolution.In the third, the dynamic control of the parameters of Evolutionary Algorithms is framed as a Reinforcement Learning problem. The control policies obtained guarantee that the optimization algorithm reaches an optimal solution within the shortest, average time
Bermell, Måns. "Identification of Problem Gambling via Recurrent Neural Networks : Predicting self-exclusion due to problem gambling within the remote gambling sector by means of recurrent neural networks." Thesis, Umeå universitet, Institutionen för fysik, 2019. http://urn.kb.se/resolve?urn=urn:nbn:se:umu:diva-159125.
Повний текст джерелаNielinger-Vakil, Carola. "Aphorism as a recurring problem in musical construction." Thesis, University of London, 1998. http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.300927.
Повний текст джерелаBenini, Fabriciu Alarcão Veiga. "Rede neural recorrente com perturbação simultânea aplicada no problema do caixeiro viajante." Universidade de São Paulo, 2008. http://www.teses.usp.br/teses/disponiveis/18/18153/tde-29042009-102601/.
Повний текст джерелаThis work proposes to solve the classic combinatorial optimization problem known as traveling salesman problem. A recurrent neural network was used in the system of optimization to search the shorter path. The structural topology linking the feedbacks of the network adopted here is known by Wang recurrent network. As learning rule to find the appropriate values of the weights was used the simultaneous perturbation with stochastic approximation. A detailed bibliographical revision on multivariable optimization with simultaneous perturbation is also described. Comparative results with other different techniques applied to the traveling salesman are still presented for validation purposes.
Venturella, Michael Anthony. "Modal Analysis of the Ice-Structure Interaction Problem." Thesis, Virginia Tech, 2008. http://hdl.handle.net/10919/31749.
Повний текст джерелаMaster of Science
Pinheiro, Tárcius Alievi. "SOLUÇÕES NÃO CLÁSSICAS PARA PROBLEMAS DA OBMEP." Universidade Federal de Santa Maria, 2013. http://repositorio.ufsm.br/handle/1/10936.
Повний текст джерелаSince 2005 the Brazilian Mathematical Olympiad of Public Schools (OBMEP) is present in many schools involving students and teachers by presenting different problems when compared to traditional models of exercises presented in textbooks. There is some special challenge in questions of OBMEP, requiring creativity in its solutions. Collecting cards and answers of the last OBMEP, in five schools in the municipality of Passo Fundo, held on June 5, 2012, we realized that exercises that involve some kind of recursive solution, there were a large numbers of errors. The objective of this work is to present alternative solutions to some problems of the OBMEP, herewith encourage high school teachers to use recurrence as a tool in building models and general solutions to mathematical problems. For this, we present some basic notes about recursive sequences, as well as a study on mathematical recurrence of first and second order, proving some basic theorems and citing some examples for better understanding of the subject. We finish, solving questions that involve recursive reasoning and that are related to OBMEP highlighting the variety of content of basic education that can be used in solving such problems.
Desde 2005, a Olimpíada Brasileira de Matemática das Escolas Públicas (OBMEP) está presente em muitas escolas envolvendo alunos e professores. Isso ocorre porque a competição apresenta problemas diferenciados quando comparados aos modelos tradicionais de exercícios contidos em livros didáticos. As questões propostas na OBMEP possuem um caráter desafiador que requer criatividade na construção das soluções. Coletamos os cartõesrespostas da última OBMEP, realizada no dia cinco de junho de 2012, em cinco escolas do município de Passo Fundo, e percebemos que exercícios envolvendo algum tipo de raciocínio recursivo apresentaram um grande número de respostas erradas. O objetivo deste trabalho é apresentar soluções alternativas para alguns problemas da OBMEP visando encorajar o professor do ensino médio a abordar e utilizar recorrência como uma ferramenta na construção de modelos e soluções gerais para problemas matemáticos. Para isso apresentamos apontamentos realizados por alguns autores a respeito de sequências recursivas. Além disso, explanamos um estudo sobre recorrência matemática de primeira e segunda ordens, no qual expomos alguns teoremas e suas respectivas demonstrações e citamos exemplos básicos para melhor compreensão do assunto. Nessa linha, selecionamos e resolvemos quatro questões que envolvem raciocínio recursivo e que estão relacionadas à OBMEP salientando a variedade de conteúdos do ensino básico que podem ser utilizados na resolução de tais problemas.
Kohler, Matthew. "What Can the Collective Action Problem Tell Us about the Recurrence of Civil War and the Long-term Stability of a Country?" Thesis, University of North Texas, 2005. https://digital.library.unt.edu/ark:/67531/metadc4955/.
Повний текст джерелаVan, Lierde Boris. "Developing Box-Pushing Behaviours Using Evolutionary Robotics." Thesis, Högskolan Dalarna, Datateknik, 2011. http://urn.kb.se/resolve?urn=urn:nbn:se:du-6250.
Повний текст джерелаWard, William. "The recurring problem of subtraction : implications and alternative solutions for teachers helping UK primary school children overcome mathematical difficulties." Thesis, University of Surrey, 2001. http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.395571.
Повний текст джерелаSouza, Márcia Erondina Dias de. "UMA PROPOSTA DE ABORDAGEM AO PROBLEMA DE FLÁVIO JOSEFO APLICADA AO ENSINO MÉDIO." Universidade Federal de Santa Maria, 2013. http://repositorio.ufsm.br/handle/1/10935.
Повний текст джерелаIn this paper, we presents a didactic sequence of activities designed for a group of students of high school, their age were about 15 and 18 years old, with the main objective to study the problem proposed by the mathematician Flávio Josefo, in mid-year 64. The legend tells that a group of rebels, including Flávio Josefo, was trapped in a cave by the enemy army. Preferring the suicide to capture, the rebels decided to form a circle and, counting over this, to kill each third person of the rest of the group. Josefo was contrary of this suicide pact therefore, together with a friend, calculated very quickly the appropriated positions that both should take in this circle in order to get out of this terrible situation. To understand this solution, we propose, at the first moment, a review about the numerical sequences, including the special cases of arithmetic and geometric. Then, we introduce some notions about the de recurrence relations and the Principle of Mathematical Induction, allowing a generalization of concepts and results already known intuitively by the student group.
Neste trabalho, apresentamos uma sequência didática de atividades elaboradas para um grupo de alunos do ensino médio, na faixa etária de 15 a 18 anos, tendo como principal objetivo estudar o problema proposto pelo matemático Flávio Josefo, nos meados do ano 64. Conta a lenda que um grupo de rebeldes, dentre eles Flávio Josefo, foi encurralado numa caverna pelo exército inimigo. Preferindo o suicídio à captura, os rebeldes decidiram formar um círculo e, contando ao longo deste, matar cada terceira pessoa restante do grupo. Josefo era contrário a este pacto suicida e, por isso, juntamente com um amigo, calculou muito rapidamente as posições adequadas que ambos deveriam tomar nesse círculo de modo a saírem ilesos desta terrível situação. Para o entendimento desta solução propomos, inicialmente, uma revisão sobre sequências numéricas, incluindo os casos especiais de progressão aritmética e geométrica. Em seguida, introduzimos algumas noções a respeito de relações de recorrência e do Princípio da Indução Matemática, permitindo uma generalização dos conceitos e resultados já conhecidos intuitivamente pelo grupo de alunos.
Flesch, Alexis. "Bandes de confiance par vraisemblance empirique : δ-méthode fonctionnelle et applications aux processus des événements récurrents". Phd thesis, Université de Franche-Comté, 2012. http://tel.archives-ouvertes.fr/tel-00955359.
Повний текст джерелаSilva, Andrea Piranhe da. "Um estudo dos zeros de polinômios ortogonais na reta real e no círculo unitário e outros polinômios relacionados." Universidade de São Paulo, 2005. http://www.teses.usp.br/teses/disponiveis/55/55134/tde-18122017-110406/.
Повний текст джерелаThe main purpose of this work is to study the behavior of the zeros of orthogonal and similar polynomials. Initially, we consider a relation between two sequences of orthogonal polynomials, where the associated measures are related to each other. We use this relation to study the monotonicity propertios of the zeros of orthogonal polynomials related with a measure obtained through a generalization of the measure associated with other sequence of orthogonal polynomials. As examples, we consider the orthogonal polynomials obtained in this way from the measures associated with the Jacobi, Laguerre and Charlier polynomials. We also consider the zeros of polynomials generated by a certain three term recurrence relation. Here, the main objective is to find bounds, in terms of the coefficients of the recurrence relation, for the regions where the zeros are located. The zeros are explored through an eigenvalue representation associated with a Hessenberg matrix. Applications to Szegõ polynomials, para-orthogonal polynomials anti polynomials with non-zero complex coefficients are considered.
Kaumanns, Franz David [Verfasser], and Hinrich [Akademischer Betreuer] Schütze. "Assessment and analysis of the applicability of recurrent neural networks to natural language understanding with a focus on the problem of coreference resolution / Franz David Kaumanns ; Betreuer: Hinrich Schütze." München : Universitätsbibliothek der Ludwig-Maximilians-Universität, 2016. http://d-nb.info/1121507999/34.
Повний текст джерелаNunes, Josiani Batista [UNESP]. "Limitantes para os zeros de polinômios gerados por uma relação de recorrência de três termos." Universidade Estadual Paulista (UNESP), 2009. http://hdl.handle.net/11449/94251.
Повний текст джерелаEste trabalho trata do estudo da localização dos zeros dos polinômios gerados por uma determinada relação de recorrência de três termos. O objetivo principal é estudar limitantes, em termos dos coeficientes da relação de recorrência, para as regiões onde os zeros estão localizados. Os zeros são explorados atravé do problema de autovalor associado a uma matriz de Hessenberg. As aplicações são consideradas para polinômios de Szego fSng, alguns polinômios para- ortogonais ½Sn(z) + S¤n (z) 1 + Sn(0) ¾ e ½Sn(z) ¡ S¤n (z) 1 ¡ Sn+1(0) ¾, especialmente quando os coeficientes de reflexão são reais. Um outro caso especial considerado são os zeros do polinômio Pn(z) = n Xm=0 bmzm, onde os coeficientes bm; para m = 0; 1; : : : ; n, são complexos e diferentes de zeros.
In this work we studied the localization the zeros of polynomials generated by a certain three term recurrence relation. The main objective is to study bounds, in terms of the coe±cients of the recurrence relation, for the regions where the zeros are located. The zeros are explored through an eigenvalue representation associated with a Hessenberg matrix. Applications are considered to Szeg}o polynomials fSng, some para-orthogonal polyno- mials ½Sn(z) + S¤n (z) 1 + Sn(0) ¾and ½Sn(z) ¡ S¤n (z) 1 ¡ Sn+1(0) ¾, especially when the re°ection coe±cients are real. As another special case, the zeros of the polynomial Pn(z) = n Xm=0 bmzm, where the non-zero complex coe±cients bm for m = 0; 1; : : : ; n, were considered.
Vandomme, Elise. "Contributions to combinatorics on words in an abelian context and covering problems in graphs." Thesis, Université Grenoble Alpes (ComUE), 2015. http://www.theses.fr/2015GRENM010/document.
Повний текст джерелаThis dissertation is divided into two (distinct but connected) parts that reflect the joint PhD. We study and we solve several questions regarding on the one hand combinatorics on words in an abelian context and on the other hand covering problems in graphs. Each particular problem is the topic of a chapter. In combinatorics on words, the first problem considered focuses on the 2-regularity of sequences in the sense of Allouche and Shallit. We prove that a sequence satisfying a certain symmetry property is 2-regular. Then we apply this theorem to show that the 2-abelian complexity functions of the Thue--Morse word and the period-doubling word are 2-regular. The computation and arguments leading to these results fit into a quite general scheme that we hope can be used again to prove additional regularity results. The second question concerns the notion of return words up to abelian equivalence, introduced by Puzynina and Zamboni. We obtain a characterization of Sturmian words with non-zero intercept in terms of the finiteness of the set of abelian return words to all prefixes. We describe this set of abelian returns for the Fibonacci word but also for the Thue-Morse word (which is not Sturmian). We investigate the relationship existing between the abelian complexity and the finiteness of this set. In graph theory, the first problem considered deals with identifying codes in graphs. These codes were introduced by Karpovsky, Chakrabarty and Levitin to model fault-diagnosis in multiprocessor systems. The ratio between the optimal size of an identifying code and the optimal size of a fractional relaxation of an identifying code is between 1 and 2 ln(|V|)+1 where V is the vertex set of the graph. We focus on vertex-transitive graphs, since we can compute the exact fractional solution for them. We exhibit infinite families, called generalized quadrangles, of vertex-transitive graphs with integer and fractional identifying codes of order |V|^k with k in {1/4,1/3,2/5}. The second problem concerns (r,a,b)-covering codes of the infinite grid already studied by Axenovich and Puzynina. We introduce the notion of constant 2-labellings of weighted graphs and study them in four particular weighted cycles. We present a method to link these labellings with covering codes. Finally, we determine the precise values of the constants a and b of any (r,a,b)-covering code of the infinite grid with |a-b|>4. This is an extension of a theorem of Axenovich
Lani, James Anthony. "Improving a Marker-Based System to Rate Assimilation of Problematic Experiences." Miami University / OhioLINK, 2003. http://rave.ohiolink.edu/etdc/view?acc_num=miami1055936557.
Повний текст джерелаJones, Benjamin Thomas. "The Past is Ever-Present: Civil War as a Dynamic Process." The Ohio State University, 2013. http://rave.ohiolink.edu/etdc/view?acc_num=osu1374173688.
Повний текст джерелаNunes, Josiani Batista. "Limitantes para os zeros de polinômios gerados por uma relação de recorrência de três termos /." São José do Rio Preto : [s.n.], 2009. http://hdl.handle.net/11449/94251.
Повний текст джерелаBanca: Alagacone Sri Ranga
Banca: Andre Piranhe da Silva
Resumo: Este trabalho trata do estudo da localização dos zeros dos polinômios gerados por uma determinada relação de recorrência de três termos. O objetivo principal é estudar limitantes, em termos dos coeficientes da relação de recorrência, para as regiões onde os zeros estão localizados. Os zeros são explorados atravé do problema de autovalor associado a uma matriz de Hessenberg. As aplicações são consideradas para polinômios de Szeg"o fSng, alguns polinômios para- ortogonais ½Sn(z) + S¤n (z) 1 + Sn(0) ¾ e ½Sn(z) ¡ S¤n (z) 1 ¡ Sn+1(0) ¾, especialmente quando os coeficientes de reflexão são reais. Um outro caso especial considerado são os zeros do polinômio Pn(z) = n Xm=0 bmzm, onde os coeficientes bm; para m = 0; 1; : : : ; n, são complexos e diferentes de zeros.
Abstract: In this work we studied the localization the zeros of polynomials generated by a certain three term recurrence relation. The main objective is to study bounds, in terms of the coe±cients of the recurrence relation, for the regions where the zeros are located. The zeros are explored through an eigenvalue representation associated with a Hessenberg matrix. Applications are considered to Szeg}o polynomials fSng, some para-orthogonal polyno- mials ½Sn(z) + S¤n (z) 1 + Sn(0) ¾and ½Sn(z) ¡ S¤n (z) 1 ¡ Sn+1(0) ¾, especially when the re°ection coe±cients are real. As another special case, the zeros of the polynomial Pn(z) = n Xm=0 bmzm, where the non-zero complex coe±cients bm for m = 0; 1; : : : ; n, were considered.
Mestre
"Solving variational inequalities and related problems using recurrent neural networks." Thesis, 2007. http://library.cuhk.edu.hk/record=b6074418.
Повний текст джерелаIn the research, the above issues are extensively explored from dynamic system perspective, which leads to the following major contributions. On one hand, many new capabilities of some existing RNNs have been revealed for solving VIs and related problems. On the other hand, several new RNNs have been invented for solving some types of these problems. The contributions are established on the following facts. First, two existing RNNs, called TLPNN and PNN, are found to be capable of solving pseudomonotone VIs and related problems with simple bound constraints. Second, many more stability results are revealed for an existing RNN, called GPNN, for solving GVIs with simple bound constraints, and it is then extended to solve linear VIs (LVIs) and generalized linear VIs (GLVIs) with polyhedron constraints. Third, a new RNN, called IDNN, is proposed for solving a special class of quadratic programming problems which features lower structural complexity compared with existing RNNs. Fourth, some local convergence results of an existing RNN, called EPNN, for nonconvex optimization are obtained, and two variants of the network by incorporating two augmented Lagrangian function techniques are proposed for seeking Karush-Kuhn-Tucker (KKT) points, especially local optima, of the problems.
Variational inequality (VI) can be viewed as a natural framework for unifying the treatment of equilibrium problems, and hence has applications across many disciplines. In addition, many typical problems are closely related to VI, including general VI (GVI), complementarity problem (CP), generalized CP (GCP) and optimization problem (OP).
Hu, Xiaolin.
"July 2007."
Adviser: Jun Wang.
Source: Dissertation Abstracts International, Volume: 69-02, Section: B, page: 1102.
Thesis (Ph.D.)--Chinese University of Hong Kong, 2007.
Includes bibliographical references (p. 193-207).
Electronic reproduction. Hong Kong : Chinese University of Hong Kong, [2012] System requirements: Adobe Acrobat Reader. Available via World Wide Web.
Electronic reproduction. [Ann Arbor, MI] : ProQuest Information and Learning, [200-] System requirements: Adobe Acrobat Reader. Available via World Wide Web.
Abstract in English and Chinese.
School code: 1307.
Wei, Zhu Zhen, and 朱振緯. "Discriminability-Optimization-Based Recurrent Fuzzy Neural Networks for Classification Problems." Thesis, 2013. http://ndltd.ncl.edu.tw/handle/20974608071614806883.
Повний текст джерела國立暨南國際大學
電機工程學系
101
The discriminative capability plays a significant role in determining classification performance. To increase the discriminative capability, this thesis proposes a Takagi–Sugeno(TS)-type maximizing-discriminability-based recurrent fuzzy network (MDRFN) which can classify highly confusable patterns. The proposed MDRFN considers minimum classification error (MCE) and minimum training error (MTE). In MCE, the weights are updated by maximizing the discrimination among different classes. In MTE, the parameter learning adopts the gradient–descent method to reduce the cost function. Therefore, the novelty of MDRFN is that it not only minimizes the cost function but maximizes the discriminative capability as well. Moreover, to enhance the “discriminability”, an enhanced discriminability recurrent fuzzy neural network (EDRFNN) was proposed. The feedback topology of the proposed EDRFNN is fully connected in order to handle temporal pattern behavior. It is constructed from structure and parameter learning. Simulations and comparisons with other recurrent fuzzy neural networks verify the performance of proposed recurrent fuzzy neural network under noisy conditions. In the experiments, other RFNs, including the singleton-type recurrent neural fuzzy network (SRNFN), TS-type RFN (TRFN), and simple RFN (SRFN), are compared. Analysis results indicate that the proposed MDRFN and EDRFNN exhibit excellent classification performance.
Zhang, Saizheng. "Recurrent neural models and related problems in natural language processing." Thèse, 2019. http://hdl.handle.net/1866/22663.
Повний текст джерелаGoyette, Kyle. "On two sequential problems : the load planning and sequencing problem and the non-normal recurrent neural network." Thesis, 2020. http://hdl.handle.net/1866/24314.
Повний текст джерелаLe travail de cette thèse est divisé en deux parties. La première partie traite du problème de planification et de séquencement des chargements de conteneurs sur des wagons, un problème opérationnel rencontré dans de nombreux terminaux ferroviaires intermodaux. Dans ce problème, les conteneurs doivent être affectés à une plate-forme sur laquelle un ou deux conteneurs seront chargés et l'ordre de chargement doit être déterminé. Ces décisions sont prises dans le but de minimiser les coûts associés à la manutention des conteneurs, ainsi que de minimiser le coût des conteneurs non chargés. La version déterministe du problème peut être formulé comme un problème de plus court chemin sur un graphe ordonné. Ce problème est difficile à résoudre en raison de la grande taille du graphe. Nous proposons une heuristique en deux étapes basée sur l'algorithme Iterative Deepening A* pour calculer des solutions au problème de planification et de séquencement de la charge dans un budget de cinq minutes. Ensuite, nous illustrons également comment un algorithme d'apprentissage Deep Q peut être utilisé pour résoudre heuristiquement le même problème. La deuxième partie de cette thèse examine les modèles séquentiels en apprentissage profond. Une stratégie récente pour contourner le problème de gradient qui explose et disparaît dans les réseaux de neurones récurrents (RNN) consiste à imposer des matrices de poids récurrentes orthogonales ou unitaires. Bien que cela assure une dynamique stable pendant l'entraînement, cela se fait au prix d'une expressivité réduite en raison de la variété limitée des transformations orthogonales. Nous proposons une paramétrisation des RNN, basée sur la décomposition de Schur, qui atténue les problèmes de gradient, tout en permettant des matrices de poids récurrentes non orthogonales dans le modèle.
Meister, Daniel [Verfasser]. "The complexity of membership problems for finite recurrent systems and minimal triangulations / von Daniel Meister." 2006. http://d-nb.info/981745369/34.
Повний текст джерелаBai, You-Chung, and 白有中. "A Recurrent Neural Fuzzy Network with Improved Particle Swarm Optimization for Solving Nonlinear System Control Problems." Thesis, 2007. http://ndltd.ncl.edu.tw/handle/20563612663583550439.
Повний текст джерела朝陽科技大學
資訊工程系碩士班
95
In this thesis, we proposed a recurrent neural fuzzy network (RNFN) with improved particle swarm optimization (IPSO) for solving nonlinear system control problem. The RNFN model is a feedforward multi-layer network which integrates traditional Takagi-Sugeno-Kang (TSK) fuzzy model and the wavelet neural networks (WNN). The recurrent property comes from feeding the internal variables, derived from membership function matched degree, back to itself. In the training strategy, we adopt supervised learning method and reinforcement learning method. In the supervised learning method consist of the improved particle swarm optimization (IPSO) and self-organizing improved particle swarm optimization (SOIPSO). The IPSO method which consists of the modified evolutionary direction operator (MEDO) and the particle swarm optimization (PSO). A MEDO combining the evolutionary direction operator (EDO) and the migration operation is also proposed. The MEDO will improve strengthen the global search solution of the evolutionary algorithm. Therefore, we use MEDO enhance PSO for solving nonlinear system control. Moreover, the SOIPSO can determine the fuzzy rules of the recurrent neural fuzzy network and tune the parameters in the RNFN automatically. The structure learning adopts several sub-swarms to constitute variable fuzzy systems and uses an elite-based structure strategy (ESS) to find suitable the number of fuzzy rules. The parameter learning is called an improved particle swarm optimization (IPSO). Although the IPSO and SOIPSO can obtain good performance in the experimental results, however in some real-world applications exact training data may be expensive or even impossible to obtain. To solve this problem, the reinforcement learning method called the reinforcement improved particle swarm optimization (R-IPSO) is proposed. In the R-IPSO, we formulate a number of time steps before failure occurs as the fitness function. The IPSO is used as a way to perform parameter learning. In the experimental results, efficiency of the proposed supervised and reinforcement learning methods are verified from these results.
Sha, Hao. "Solving Prediction Problems from Temporal Event Data on Networks." Thesis, 2021. http://dx.doi.org/10.7912/C2/46.
Повний текст джерелаMany complex processes can be viewed as sequential events on a network. In this thesis, we study the interplay between a network and the event sequences on it. We first focus on predicting events on a known network. Examples of such include: modeling retweet cascades, forecasting earthquakes, and tracing the source of a pandemic. In specific, given the network structure, we solve two types of problems - (1) forecasting future events based on the historical events, and (2) identifying the initial event(s) based on some later observations of the dynamics. The inverse problem of inferring the unknown network topology or links, based on the events, is also of great important. Examples along this line include: constructing influence networks among Twitter users from their tweets, soliciting new members to join an event based on their participation history, and recommending positions for job seekers according to their work experience. Following this direction, we study two types of problems - (1) recovering influence networks, and (2) predicting links between a node and a group of nodes, from event sequences.
Huang, Ching-I., and 黃瀞儀. "Through the Theory of Risk Society to Analyze the problems and Strategies in Recurrent Education in Higher Education in Taiwan." Thesis, 2006. http://ndltd.ncl.edu.tw/handle/48616599713836624850.
Повний текст джерела國立中正大學
成人及繼續教育所
94
The research focuses on“risk society”and “the development of macro dimension, meso dimension and micro dimension in recurrent education in higher education in Taiwan ”. And also, the research bases on “risk society”which is advocated by Beck and Giddens. Through the theory to analyze the problems and strategies in recurrent education in higher education in Taiwan. The background of the research is about the transition from Industrial Society to Risk Society. Accordingly, the policies, curriculums and individual learners in the recurrent education in higher education must face many risks and problems. Taken altogether, the purposes of the research is “examine the theory of Risk Society”, “understand the development and comprehension of recurrent education in higher education”, “through the theory of risk society to analyze the risks of the recurrent education in higher education in Taiwan ”and “establish the strategies to solve the risks”. The research uses the method of Theory Analysis and with some discussion and analysis, came to the following conclusions. 1.The theory of Risk Society includes the comprehension of“risk”, “individualization”, and “reflexive modernization”. 2.The theory of Risk Society emphasize “learning reason”, “life narrative ”, and “reflexive learning”. 3.The original of the recurrent education is from Sweden, and the essence emphasize the equality of educational opportunity and learning right. 4.The recurrent education emphasize the learning model of education-work-leisure. 5.The development of the recurrent education in higher education in Taiwan includes protecting the learning right, opening the gat of the recurrent education in higher education, paid education leave, and accreditation of learning achievement. 6.The development of the recurrent education in higher education in Taiwan can generalize 10 phenomenon in the dimensions of policies, curriculums and individual learners. 7.The development of the recurrent education in higher education in Taiwan can generalize 10 risks in the dimensions of policies, curriculums and individual learners. 8.Finally, we can develop 7 strategies to keep off the risks in recurrent education in higher education in Taiwan. Key words:Risk, Risk society, Recurrent Education in Higher Education, Individualization, Reflexive Modernization
(11048391), Hao Sha. "SOLVING PREDICTION PROBLEMS FROM TEMPORAL EVENT DATA ON NETWORKS." Thesis, 2021.
Знайти повний текст джерелаMany complex processes can be viewed as sequential events on a network. In this thesis, we study the interplay between a network and the event sequences on it. We first focus on predicting events on a known network. Examples of such include: modeling retweet cascades, forecasting earthquakes, and tracing the source of a pandemic. In specific, given the network structure, we solve two types of problems - (1) forecasting future events based on the historical events, and (2) identifying the initial event(s) based on some later observations of the dynamics. The inverse problem of inferring the unknown network topology or links, based on the events, is also of great important. Examples along this line include: constructing influence networks among Twitter users from their tweets, soliciting new members to join an event based on their participation history, and recommending positions for job seekers according to their work experience. Following this direction, we study two types of problems - (1) recovering influence networks, and (2) predicting links between a node and a group of nodes, from event sequences.
Chen, Muye. "Vanishing Gradient Problem in Training Neural Networks." Thesis, 2022. http://hdl.handle.net/1885/268662.
Повний текст джерелаChin, Tai-Lin, and 金台齡. "On the Time Scheduling Problem of Uniform Recurrence Equations." Thesis, 1997. http://ndltd.ncl.edu.tw/handle/44704529731090887534.
Повний текст джерела國立交通大學
資訊工程學系
85
In this thesis, we study a series of hyperplane-based time scheduling methods for uniform recurrence equations. The studied schedules are linear schedule, uniform affine schedule, affine schedule, and multi-dimensional schedule.In addition, a schedule named as multi-dimensional affine schedule is studied.It improves the execution time of the multi-dimensional schedule. All these scheduling problems are formulated as linear programs. Time scheduling problem is a very important problem on parallel cmputing field. It plays an important role in the designs of parallel processor array and parallel compilers. Besides, for illustration, these scheduling methods are implemented for some real examples, including convolution, selection sort, 2D filter and a complex example of uniform recurrence equations. From these illustrative examples, the quality of various time scheduling methods can be explored. Moreover, the importance of the time scheduling problem in parallel algorithms can be seen.
Kanuparthi, Bhargav. "Towards better understanding and improving optimization in recurrent neural networks." Thesis, 2020. http://hdl.handle.net/1866/24319.
Повний текст джерелаLes réseaux de neurones récurrents (RNN) sont connus pour leur problème de gradient d'explosion et de disparition notoire (EVGP). Ce problème devient plus évident dans les tâches où les informations nécessaires pour les résoudre correctement existent sur de longues échelles de temps, car il empêche les composants de gradient importants de se propager correctement sur un grand nombre d'étapes. Les articles écrits dans ce travail formalise la propagation du gradient dans les RNN paramétriques et semi-paramétriques pour mieux comprendre la source de ce problème. Le premier article présente un algorithme stochastique simple (h-detach) spécifique à l'optimisation LSTM et visant à résoudre le problème EVGP. En utilisant cela, nous montrons des améliorations significatives par rapport au LSTM vanille en termes de vitesse de convergence, de robustesse au taux d'amorçage et d'apprentissage, et de généralisation sur divers ensembles de données de référence. Le prochain article se concentre sur les RNN semi-paramétriques et les réseaux auto-attentifs. L'auto-attention fournit un moyen par lequel un système peut accéder dynamiquement aux états passés (stockés en mémoire), ce qui aide à atténuer la disparition des gradients. Bien qu'utile, il est difficile à mettre à l'échelle car la taille du graphe de calcul augmente de manière quadratique avec le nombre de pas de temps impliqués. Dans l'article, nous décrivons un mécanisme de criblage de pertinence, inspiré par le processus cognitif de consolidation de la mémoire, qui permet une utilisation évolutive de l'auto-attention clairsemée avec récurrence tout en assurant une bonne propagation du gradient.
Lemańczyk, Michał. "Recurrence of stochastic processes in some concentration of measure and entropy problems." Doctoral thesis, 2022. https://depotuw.ceon.pl/handle/item/4158.
Повний текст джерелаNiech X = (Xi)i∈Z, gdzie Xi ∈ X a X jest (mierzaln¡) przestrzeni¡ stanów, b¦dzie procesem stochasty- cznym. Niniejsza rozprawa doktorska koncentruje si¦ na procesach czasów powrotu R = (Ri)i∈Z kolejnych powrotów Xi do A oraz ich roli zarówno w teorii prawdopodobie«stwa, jak i w teorii ergody- cznej. Przypomnijmy, »e dla danego podzbioru A ⊂ X odpowiadaj¡cy mu proces czasów powrotu jest zde niowany jako Ri = inf{j ≥ 0 : Xj ∈ A}, i = 0, inf{j > Ri−1 : Xj ∈ A}, i ≥ 1, sup{j < Ri+1 : Xj ∈ A}, i ≤ −1. (1.1) Gªównym rezultatem rozprawy w teorii prawdopodobie«stwa jest dowód nierówno±ci Bernsteina dla funkcjonaªów addytywnych ogólnych, niekoniecznie silnie aperiodycznych, ªa«cuchów Markowa, co daje odpowied¹ na pytanie sformuªowane w pracy [1] (patrz [A3]). Dowodzimy równie» pewnej nowej wersji nierówno±ci Bernsteina dla 1-zale»nych procesów (klasa ta jest silnie zwi¡zana z ªa«cuchami Markowa dzi¦ki tzw. technice regeneracji). Gªówne rezultaty w teorii ergodycznej dotycz¡ dokªadnych wzorów, b¡d¹ nierówno±ci, zwi¡zanych z entropi¡ (ang. entropy rate) punktowego iloczynu procesów (patrz [A1]). Staj¡ si¦ one narz¦dziem do rozwi¡zania kilku otwartych problemów. Podajemy nowy, jawny wzór na ci±nienie topologiczne ukªadów BBB-wolnych oraz, w pewnych przypadkach, dowodzimy jedyno±ci stanów równowagi dla ukªadu wyznaczonego przez BBB (co rozszerza rezultaty o wewn¦trznej ergodyczno±ci udowodnione w [3, 2]). Odpowiadamy na pytanie postawione w [3] o braku wªasno±ci Gibbsa dla miary o maksymalnej entropii (patrz [A2]). W ko«cu, odpowiadamy na kilka pyta« doty- cz¡cych entropii ukªadów BBB-wolnych z pracy [2] (patrz [A1]). Cz¦±¢ rezultatów rozprawy jest nowa, pozostaªe rezultaty pochodz¡ z nast¦puj¡cych trzech artykuªów: [A1] J. Kuªaga-Przymus and M.D. Lema«czyk. Entropy rate of product of independent processes. Preprint: arXiv:2004.07648, 2020. [A2] J. Kuªaga-Przymus and M.D. Lema«czyk. Hereditary subshifts whose measure of maximal en- tropy has no Gibbs property. To appear in Colloquium Mathematicum, arXiv:2004.07643, 2020. [A3] M.D. Lema«czyk. General Bernstein-like inequality for additive functionals of Markov chains. Journal of Theoretical Probability, 2020.
Liao, Jian-Wei, and 廖建維. "An Ensemble Learning Approach for Data Stream with Concept Drift and Recurring Class Problem." Thesis, 2014. http://ndltd.ncl.edu.tw/handle/39769659102476301628.
Повний текст джерела國立臺灣科技大學
資訊工程系
102
Recently, concept drift has become an important issue while analyzing non-stationary distribution data in data mining. For example, data streams carry a characteristic that data vary by time, and there is probably concept drift in this type of data. Concept drifts can be categorized into sudden and gradual concept drifts in brief. Most of researches concentrate on solving one type of concept drift since the type of concept drift is usually difficult to be identified. Moreover, there is an important issue in the data stream called recurring class problem which is caused by concept drift. This problem often occurs in the non-stationary environment. In light of these reasons, we propose two new weighting mechanisms to base models in the ensemble and new arriving data in order to adapt to current concept quickly. The experimental results show that our method is not only stable enough on the datasets with gradual concept drifts but also flexible to adapt to sudden concept drifts and the recurring class problem.
Senna, Daniel Lordêllo. "O Estatuto do Árbitro Numa Perspectiva De Direito Comparado Entre Portugal e Brasil." Master's thesis, 2017. http://hdl.handle.net/10316/84014.
Повний текст джерелаO Estatuto do Árbitro Numa Perspectiva De Direito Comparado Entre Portugal e Brasil.O árbitro é figura central do instituto da arbitragem, e, neste trabalho, procurou-se enfatizar a importância da sua atuação para o crescimento desta forma alternativa de solução de litígios, com especial ênfase nos ordenamentos português e brasileiro.Verificou-se uma série de problemas e dificuldades às quais estão os árbitros submetidos ao cumprirem a sua função, especialmente aqueles relacionados à questão de quem pode exercer o encargo, os deveres inerentes à investidura e principalmente os atributos da independência e imparcialidade.Na mesma esteira, não se deixou de analisar os direitos dos árbitros e demais partes envolvidas na arbitragem. Questões controvertidas na jurisprudência foram analisadas, especialmente aquelas atinentes à recusa do árbitro e o dever de revelação, problemas que claramente predominam entre as intempéries às quais está exposta a arbitragem.O presente trabalho contém, ainda, uma sugestão de alteração nas legislações portuguesa e brasileira, para que se passe a exigir a presença de ao menos um jurista na constituição do tribunal arbitral, com o objetivo de fazer reduzir o número de incidentes trazidos por algum desconhecimento da matéria legal dos árbitros e seu estatuto.Frise-se que tal exigência já existe na legislação espanhola e em diversos outros países hispano-americanos. Resta claro que esta proposta passa longe de ser inatacável ou livre de críticas, posto que boa parte da doutrina defende que tal alteração seria antidemocrática, corporativista e protecionista.Contudo, observa-se que os problemas que envolvem a arbitragem no Brasil e em Portugal, estão, em grande medida, relacionados à falta de conhecimento, por parte dos árbitros não juristas, de premissas básicas e basilares do direito, nomeadamente questões que envolvem a independência e imparcialidade como condição precípua para o exercício da função, afinal, tais atributos são fundamento da legitimidade do poder de julgar. Também importante é a discussão que se trava na jurisprudência portuguesa acerca do grau de fundamentação da sentença arbitral. Há acórdãos recentes em que se percebe a absoluta discrepância entre os entendimentos acerca da matéria.Aqui se defendeu o entendimento de que há rigor excessivo por parte dos juízes dos tribunais estaduais e que tal exigência acaba por abrir flancos indesejáveis a ações anulatórias que, em última análise, enfraquecem o instituto da arbitragem.
The statute of the arbitrator in a comparative law perspective between Portugal and Brazil.The arbitrator is a central figure of the arbitration institute and, in this work, we tried to emphasize the importance of its action for the growth of this alternative form of dispute settlement, with special emphasis on Portuguese and Brazilian law.There have been a number of problems and difficulties to which the arbitrators are subjected in carrying out their duties, especially those related to the question of who can perform the task, the duties inherent to the investiture, and especially the attributes of independence and impartiality.In the same vein, the rights of the arbitrators and other parties involved in the arbitration were also examined. Controversial issues in the case-law have been examined, especially those concerning refusal by the arbitrator and the duty to disclose, problems which clearly prevail among the weather to which arbitration is exposed.The present work also contains a suggestion of alteration in Portuguese and Brazilian legislation, so that it is necessary to have at least one jurist present in the constitution of the arbitral tribunal, in order to reduce the number of incidents brought about by some unknown Of the legal matter of the arbitrators and their status.It is frightened that such a requirement already exists in Spanish law and in several other Spanish-American countries. It remains clear that this proposal is far from being unassailable or free from criticism, since much of the doctrine holds that such a change would be undemocratic, corporatist and protectionist.However, it is observed that the problems involving arbitration in Brazil and Portugal are, to a large extent, related to the lack of knowledge by non-legal arbitrators of basic and basic premises of law, namely issues involving Independence and impartiality as the precondition for the exercise of the function, after all, such attributes are the basis of the legitimacy of the power to judge.Also important is the discussion in Portuguese jurisprudence about the degree of justification of the arbitration award. There are recent judgments in which the absolute discrepancy between the understandings about the matter is perceived.Here it was defended the understanding that there is excessive rigor on the part of the judges of the state courts and that this requirement ends up opening undesirable flanks to annulatory actions that, in the end, weaken the institute of the arbitration.
Beauregard, Julie. "Trajectoires et profils des adolescents placés en centre de réadaptation présentant des troubles de comportement sérieux." Thèse, 2018. http://hdl.handle.net/1866/21339.
Повний текст джерела